Preliminary study to examine the utility of using foot burn or hock burn to assess aspects of housing conditions for broiler chicken

1. Eleven broiler chicken farms, representing 4 production system types, were visited during the last 5 d of the flock cycle: bird and flock details were recorded. Litter friability was assessed at 9 sites within the house, atmospheric ammonia was measured at three sites and bird cleanliness was assessed on a numerical rating scale. 2. For these flocks, hock burn, foot burn and breast burn were measured at the processing plant by standardised assessors. 3. Significant correlations were identified between the percentage of birds with foot burn and average litter score, average house ammonia concentrations and feather score. 4. No correlation was found between the percentage of birds with hock burn or breast burn and average litter scores, average ammonia concentrations or feather score. 5. No correlation was found between stocking density and foot burn, hock burn or breast burn.6. If confirmed, these findings may have implications for the draft EU Broiler Directive, for which it is proposed that permitted stocking density on farm may be determined by the incidence and severity of contact dermatitis measured on plant.     

Factors affecting the prevalence of foot pad dermatitis, hock burn and breast burn in broiler chicken

1. Standardised data on flock husbandry were recorded on 149 broiler farms during the 4 d prior to slaughter. 2. Birds were examined at the slaughterhouse for contact dermatitis lesions. Foot pad dermatitis score (FPDS) and hock burn score (HBS) were measured on five point scales. Carcase rejection data were also collected. 3. The mean percentage of birds in each flock with: moderate or severe foot lesions was 11.1% (range 0-71.5%); moderate or severe hock burn was 1.3% (range 0-33.3%); and, breast burn was 0.02%. 4. A general linear model was developed to examine factors associated with mean flock FPDS. Assuming a linear relationship, within the range of data collected and with all other factors remaining the same, every 1% increase in the proportion of Genotype A birds in the flock was associated with an increase in mean FPDS of 0.003, every one-point increase in litter score was associated with a 0.326 increase in mean FPDS and every one-point increase in flock mean HBS was associated with a 0.411 increase in mean FPDS. Flock mean FPDS was associated with feed supplier and was higher in winter. 5. The general linear model developed for flock mean HBS, found that every one-point increase in mean FPDS increased mean HBS by 0.090, every one-point increase in litter score increased HBS by 0.119 and, every 1% increase in small/emaciated birds decreased mean HBS by 0.333. Reduced HBS was also associated with increased final litter depth, younger slaughter age and an increased percentage of dietary wheat. For every 1% increase in Genotype A birds, a decrease in flock mean HBS of 0.003 would be expected. 6. An effect of hatchery was also identified.     

A comparison survey of organic and conventional broiler chickens for infectious agents affecting health and food safety

The purpose of the present cross-sectional study was to evaluate the health status of organic broiler chickens and the contamination rate with Salmonella and Campylobacter in organic broiler production in Belgium. The broilers were screened for antibodies against routinely monitored poultry diseases at 1 day old and at slaughter. Fecal examination for the presence of worm eggs was done at slaughter. Bacteriological examination for the detection of Salmonella and Campylobacter was performed at day 1, week 2, week 4, week 7, week 10, and slaughter. Conventional broilers of the same poultry integration and reared in the same geographic area were also screened and served as reference. Serologic data indicated lower antibody titers against infectious bronchitis and Newcastle disease in organic flocks. No significant differences could be found in prevalence of Salmonella between organic and conventional broilers at slaughter. In contrast, Campylobacter infections at slaughter were significantly higher in organic flocks. Organic flocks most probably become infected with Campylobacter between week 7 and week 10. Worm eggs were found in neither the organic flocks nor the conventional flocks. In conclusion, there are indications that the respiratory health status is better in organic broilers but that organic flocks are more often infected with Campylobacter than are conventional flocks.     

A 1-year epidemiological study of campylobacters in 18 Swedish chicken farms

Broiler chickens are often intestinal carriers of Campylobacter. During processing, Campylobacter may be spread over the carcass. Thus, undercooked chicken meat, or other foods contaminated by raw chicken can act as a source of infection to humans. This study was conducted to identify risk factors for chicken flocks being colonized with Campylobacter. Eighteen chicken farms with altogether 62 chicken compartments were studied for 1 year with visits during each growing period and sampling of chicken caecal contents at slaughter. Four to six subsequent flocks were raised in each compartment during the study. A detailed questionnaire was used to record farm parameters such as building materials, feed and water equipment, hygiene and management routines. Campylobacter prevalence varied between farms, between growing periods within the farms and also during the year, with lowest prevalence during the spring. Campylobacters were isolated from 27% out of 287 flocks. Only two farms were negative at all samplings. Often the flock following a positive flock in a compartment was negative, indicating that normal cleaning and disinfecting routines are sufficient for eliminating the bacteria from the house. Usually only one serotype was found in each positive flock. Campylobacter occurrence increased with the age of the chickens at slaughter, and also with flock size.

Univariable chi-square tests were done of the association between possible risk factors and Campylobacter prevalence. Factors associated with higher Campylobacter prevalence in flocks were lack of or diffuse hygiene barriers, increasing flock size, increasing age at slaughter, short vs. long empty periods, wet litter beds, other poultry nearby or staff handling other poultry, flocks divided before slaughter, staff loading to slaughter at several farms and occurrence of mice. Under Swedish conditions, water does not seem to be a source of infection for chickens. Origin and handling of day-old chickens, feed additives, houses and litter were not associated with higher Campylobacter prevalence.     

Epidemiology and significance of chicken infectious anemia virus infections in broilers and broiler parents under nonvaccinated European circumstances

A serologic survey in unvaccinated broiler parent and broiler progeny flocks demonstrated seroconversion against chicken infecrious anemia virus (CIAV) in all parent flocks before or around point of lay and in 38% of the broiler flocks examined at slaughter age. The presence of CIAV antibodies at slaughter of broilers was positively correlated with slaughterhouse condemnation rates. Results indicate that CIAV infections are highly prevalent in both broiler parent and broiler flocks and that CIAV infections in broilers are associated with increased slaughterhouse condemnation.     

Campylobacter Cross‐Contamination of Chicken Products at an Abattoir

Consumption of raw or undercooked poultry products contaminated with Campylobacter has been identified as a risk factor for human campylobacteriosis. We determined whether slaughtering of Campylobacter‐positive flocks was associated with contamination of chicken products derived from Campylobacter‐negative flocks slaughtered at the same abattoir. The presence of Campylobacter was investigated in 22 broiler farms 1 week prior to slaughter and in one abattoir on nine separate slaughter days. A total of 600 bulk packed chicken products were tested, with 198 (33.0%) of the products found to be Campylobacter positive. Of the 350 chicken products originating from Campylobacter‐positive flocks, 180 (51.1%) were contaminated with the bacteria. In contrast, only 18 (7.2%) of 250 chicken products derived from Campylobacter‐negative flocks were contaminated. In 14 of these 18 products, the Campylobacter isolates were identical to isolates obtained from the flock slaughtered immediately prior to the Campylobacter‐negative flock. Notably, on 4/6 slaughter days, Campylobacter‐negative flocks were slaughtered prior to the positive flocks, and Campylobacter was absent from all chicken products originating from the negative flocks. These results suggest that implementation of logistic slaughter (where Campylobacter‐negative flocks are slaughter first) significantly decreases the prevalence of Campylobacter‐positive chicken products.     

National surveillance of Campylobacter in broilers at slaughter in Denmark in 1998

A surveillance study for thermophilic Campylobacter spp. in broiler flocks was carried out for the year 1998 in Denmark. The study included examinations of 4286 broiler flocks comprising samples from 57,000 birds. Overall, a flock prevalence of 46.0% was recorded. The species distribution was Campylobacter jejuni 86%, Campylobacter coli 11%, Campylobacter lari 1%, other not further diagnosed species 2%. The prevalence was significantly higher in the period from June to October (3.2 < odds ratio [OR] <1.8, P < 0.0002) and was significantly associated with abattoir (OR < 2.8, P < 0.0001) and the length of the period the broiler houses were left empty between flocks (download period; 6 days or more) (OR = 1.6, P < 0.0198). No association between Campylobacter colonization and the age at slaughter was found. Separating the flocks into batches for slaughter elevated the flock prevalence from 0.41 after the first batch had been slaughtered to 0.46 after all batches had been slaughtered.     

Prevalence and control of foot-pad dermatitis in broilers in Sweden

1. A cross-sectional observational study was carried out to investigate spatial, temporal and management aspects of foot-pad dermatitis in Swedish broilers. The efficacy of a control programme was evaluated. 2. Flock information on producer, breed, foodstuff manufacturer, region, abattoir, date of slaughter, age at slaughter, planned and actual stocking density was recorded. A total of 6988 flocks, representing approximately 110 million broilers, was examined. A total of 175 broiler producers from 15 geographical regions is represented. 3. The total foot-pad score per flock ranged from 0 to 200, with a mean of 34.7 and a standard deviation of 40.9. 4. Simple linear regression analysis showed that the mean total score on a weekly basis decreased (P 0.001) over time, mainly because of a decrease in the prevalence of severe foot-pad lesions. 5. There was an association (P 0.05) between slaughterhouse, foodstuff supplier and geographic region respectively and the mean total foot pad score of the flocks slaughtered.     

Distribution of Campylobacter jejuni isolates from turkey farms and different stages at slaughter using pulsed-field gel electrophoresis and flaA-short variable region sequencing

The aim of this study was to assess the diversity of thermotolerant Campylobacter spp. isolated from turkey flocks at six rearing farms 1-2 weeks prior to slaughter (360 faecal swab samples) and from 11 different stages at the slaughterhouse (636 caecal, environmental, neck skin and meat samples). A total of 121 Campylobacter isolates were identified to species level using a multiplex PCR assay and were typed by pulsed-field gel electrophoresis (PFGE) and flaA-short variable region (SVR) sequencing. All Campylobacter isolates were identified as Campylobacter jejuni. PFGE analysis with KpnI restriction enzyme resulted in 11 PFGE types (I-XI) and flaA SVR typing yielded in nine flaA-SVR alleles. The Campylobacter-positive turkey flocks A, C and E were colonized by a limited number of Campylobacter clones at the farm and slaughter. The present study confirms the traceability of flock-specific strains (PFGE types I, V and IX; flaA types 21, 36 and 161) from the farm along the entire processing line to meat cuts. It seems that stress factors such as high temperature of the defeathering water (54-56 °C), drying of the carcass skin during air chilling (24 h at 2 °C), and oxygen in the air could not eliminate Campylobacter completely. Campylobacter-negative flocks became contaminated during processing by the same subtypes of Campylobacter introduced into the slaughter house by preceeding positive flocks even if they were slaughtered on subsequent days. Proper and efficient cleaning and disinfection of slaughter and processing premises are needed to avoid cross-contamination, especially in countries with a low prevalence of Campylobacter spp. The majority of flaA SVR alleles displayed a distinct association with a specific PFGE type. However, a linear relationship for all strains among both typing methods could not be established. To specify genetic relatedness of strains, a combination of different genotyping methods, is needed.     

Antibody prevalence of low-pathogenicity avian influenza and evaluation of management practices in Minnesota backyard poultry flocks

Low‐pathogenicity avian influenza (LPAI) viruses have caused illness in poultry and humans with poultry contact. To determine whether there is evidence of exposure to avian influenza viruses (AIV) among backyard poultry in Minnesota and their human caretakers, 150 flocks of backyard birds were sampled for antibodies to AIV from August 2007 through December 2008. One hundred flocks were tested through routine slaughter surveillance by the Minnesota Board of Animal Health and an additional 50 flocks were contacted and sampled by study investigators. Blood was collected from 10 to 13 birds from each flock and a survey of biosecurity and management practices was administered to the flock owner. Blood samples were tested by agar gel immunodiffusion (AGID) for influenza A antibodies. Tested flocks had a median flock size of 100 birds (range: 12–800 birds), and were most commonly owned for meat for personal use (81% of respondents), fun or hobby (58%) and eggs for personal use (56%). Although 7% of flock owners reported that their birds had shown respiratory signs in the previous 3 months, only 1 of 150 flocks tested positive for influenza by AGID. Antibodies to LPAI H6N1 were detected in the positive flock. The owner of the positive flock did not have antibodies to H6 or other common AIV. Based on the findings of this study, the risk of transmission of LPAI viruses from backyard poultry to owners in Minnesota appears to be low under current conditions and management practices.     

Dynamics of Campylobacter spp. spread investigated in 14 broiler flocks in Switzerland

Ten conventional and four extensive outdoor broiler flocks, distributed over nine farms, were investigated twice per week during a 35-58-day rearing period to observe the dynamics of Campylobacter spp. spread within these flocks. Strains isolated during this period were genotyped by restriction fragment length polymorphism analysis of the flaA gene and macrorestriction profiling with pulsed field gel electrophoresis. A total of 4112 samples were collected; 157 (3.8%) of these samples were Campylobacter positive, with all C. jejuni. The positive samples were distributed over three conventional and two extensive outdoor flocks on five farms. These five positive flocks were colonized from the fifth to the seventh week of age and remained colonized until slaughter. Each of the flocks showed a flock-specific genotype of Campylobacter that predominated until slaughter. Presuming different ways of entry, a combination of this fact and the observed dynamics of C. jejuni spread within the flocks indicates that a single source from the environment may have been responsible for the colonization of each flock. These conclusions may serve to further develop combat strategies at farm level.     

Chicken welfare as indicated by lesions on carcases in supermarkets

1. The extent of visible lesions in British broiler carcases was investigated in order to discover the extent of any poor welfare. It is well established that hock burn and some other injuries can occur during housing of broiler chickens, whilst other injuries can occur during handling. 2. Three hundred and eighty-four whole conventionally reared British Farm Standard Grade A chickens were scrutinised in supermarkets to assess the frequency of 15 kinds of lesions. Six lesions were analysed histopathologically and this showed that hock burn would have occurred several days pre-mortem, red alular and wrist bruise were very likely to have occurred pre-mortem, whilst red parson's nose and scratches were likely to have occurred pre-mortem. 3. Many broiler carcases in the supermarkets had dermal lesions: 0.82 had hock burns, 0.45 a red parson's nose, 0.32 a red alular, 0.21 wrist bruises and 0.21 had at least one scratch on a leg. The frequency of the larger hock burns increased with body weight. 'Organic' chickens had half as many hock burns as conventionally reared broilers, perhaps because of differences in litter quality or leg strength. 4. The sample studied indicates that painful lesions which would result in poor welfare are frequent in broilers slaughtered in the UK. The Grade A chickens observed exclude birds with obvious visible defects because these birds would have had the blemishes removed and the carcase would have been portioned. Hence the frequencies of lesions in farmed birds would be higher than those reported here.     

Broiler chicken health, welfare and fluctuating asymmetry in organic versus conventional production systems

The aim of this study was to test three predictions: (1) that a combination of animal-based parameters indicates better health and welfare in broilers from organic versus conventional farms, (2) that broilers from organic farms show less fluctuating asymmetry (FA) than broilers from conventional farms, and (3) that, at the level of the individual bird, the relationship between FA and welfare is negative and strongest in conventional broilers. On 140 slaughter-age birds randomly selected from seven organic and seven conventional flocks in Belgium we measured standardised FA and five conventional animal-based welfare indicators (tonic immobility duration, latency-to-lie, and condition of the foot pad, hock, and breast). The caeca from the birds from four organic and four conventional flocks were removed for assessing the presence of two bacterial pathogens (Salmonella and Campylobacter) and the concentration of the health-promoting lactic acid bacteria. Finally, the blood serum concentration of the acute phase protein alpha-1-acid glycoprotein (AGP), a non-specific indicator of immunological stress, was determined. Salmonella was found in the caeca content of two (2.5%) birds only (both from the same conventional farm), whereas 44 (55.7%) birds were infected with Campylobacter. The prevalence of Campylobacter, the concentration of lactic acid bacteria, the duration of tonic immobility, and the condition of the breast and foot pad did not differ between both production systems. Apart from a higher concentration of AGP, organic birds had better scores for hock condition and a longer latency-to-lie indicating better leg health. In addition, organic birds scored better on the aggregated welfare index (i.e. the average of the five standardised welfare indicators). As both production systems differed in many aspects (e.g. slower-growing genotypes, slaughter age, feed, stocking density, group size) it is impossible to assign differences in welfare/health indicators to a single factor. Whatever the causes may be our findings suggest that, despite the potentially elevated risk of immunological challenge, broiler chicken welfare is generally superior in organic farms as compared with conventional farms in Belgium. Regarding the validity of FA as welfare indicator, the prediction of lower FA in the population with highest aggregated welfare score was confirmed but, at the level of the individual, no associations between FA and the aggregated welfare index were found irrespective of whether data from organic and conventional broilers were analysed separately or combined.     

Economic effects of subclinical chicken anemia agent infection in broiler chickens

To evaluate the economic effects of subclinical chicken anemia agent (CAA) infection on broiler performance, clinically normal broiler flocks were grouped into two categories: A) flocks in which none of 10 birds sampled at slaughter had antibody to CAA, and B) flocks in which six or more of 10 similarly sampled birds had CAA antibody. Production and performance parameters of 25 flocks in each category were compared. No statistically significant differences were found between category A and category B flocks in major production parameters such as sex, feed manufacturer, type of litter, stocking density, and age at slaughter. However, category A flocks achieved 13% (P less than 0.05) greater net income per 1000 birds, 2.0% better feed-conversion ratio (P less than 0.05), and 2.5% (P less than 0.05) greater average weight per bird than category B flocks. No significant differences in hockburn bonus per 1000 birds (a bonus payable to growers whose broilers have the lowest prevalence of contact dermatitis lesions on the hocks) and mortality were found between category A and category B flocks. These results show that subclinical CAA infection has a substantial, statistically significant effect on commercial broiler performance and profitability.     

Characterization of Small-Scale Antibiotic-Free Broiler Production in Minnesota

The purpose of this study was to characterize the production practices of small antibiotic-free flock producers and identify the possible points for introduction of food-borne pathogenic bacteria. These data will be used to develop appropriate extension programs to improve the level of biosecurity on such farms. Minnesota poultry producers who raised flocks of meat-type chickens without antibiotics were interviewed either on their farms or via the telephone. A wide diversity of housing and management styles were disclosed. Key areas identified for extension programs include feed and pasture choice, waste disposal, feed withdrawal before slaughter, and marketing.     

Predicted quantitative effect of logistic slaughter on microbial prevalence

Logistic slaughter is an intervention measure intended to reduce cross-contamination during slaughter by slaughtering contaminated units (=(groups of) animals) last. This paper describes a simple mathematical model which predicts the prevalences of contaminated units after logistic and random-order slaughter. The effect of logistic slaughter is the difference between these prevalences. The model assumes that uncontaminated units can become contaminated by contaminated units that were slaughtered before them; the contributions of contaminated units are independent. It also assumes that a slaughterhouse is uncontaminated at the start of the day and that a unit that is contaminated before slaughter also is contaminated after slaughter. The model was analysed using numerical simulations; for a selection of cases, analytical formulas can be derived and are presented. Contamination of broiler flocks with Salmonella was used as a case study. Even for this simple model, data availability is a problem leading to uncertain parameter estimates. An average cross-contamination scenario predicts that the beneficial effect of logistic slaughter is as low as 9.1%, which casts doubt on its usefulness as an intervention measure. The case study produced these general model results: the effect of logistic slaughter increases with the probability of cross-contamination between units; with the length of the slaughter queue; and with sensitivity (the probability of a positive test from a unit contaminated at the start of slaughter). However, the effect is small if the prevalence of contaminated units before slaughter is low or high.     

Hierarchical and spatial analyses of pneumonia-lesion prevalence at slaughter in New Zealand lambs

We recorded lesions of moderate-to-severe pneumonia (>or=10% lung surface area affected; "pneumonia") in 1,899,556 lambs submitted to three New Zealand abbatoirs between December 2000 and September 2001. The average prevalence of pneumonia ranged between 7 and 13%, by abbatoir. We ran a two-level mixed-effects binomial logistic-regression model with the prevalence of pneumonia as the outcome, and adjusting for abbatoir and month. The intracluster correlations for batch (slaughter lambs from the same farm sent at the same time) and farm were 31.3 and 12.4%, respectively. (We also noted threefold differences in odds across abbatoirs, and >30-fold differences among slaughter months.) Case flocks (those in the upper quintile of pneumonia prevalence) generally were not clustered in the spatial incidence-risk analysis (after adjusting for flock-level and batch level effects, and as compared to flocks in the lower two quintiles). We therefore concluded that the risk of moderate-to-severe pneumonia-lesion prevalence detectable at slaughter of lambs was determined at the flock and batch level, rather than at the spatial level.     

Causes of mortality in commercial organic layers in Denmark

A longitudinal study investigated the courses of mortality in commercial free-range organic layer flocks in Denmark. In total, 15 organic egg-producing flocks from 11 farms were randomly selected among 80 farms registered in Denmark. Four farms with confined egg production on deep litter were included for comparison. Flock sizes ranged from 2260 to 5940 layers. The flocks were monitored from introduction to the layer farm until slaughter. Flock mortalities ranged from approximately 2% to 91%, with a mean of 20.8% for organic flocks compared with 7% for confined flocks on deep litter. In total, 4608 layers were submitted for postmortem examination, representing > 40% of all the dead layers in the investigated flocks. Outbreaks of erysipelas (Erysipelothrix rhusiopathiae) and fowl cholera (Pasteurella multocida) were observed in two and three organic flocks, respectively. The mortality rate reached 91% in one organic flock dually affected by erysipelas and fowl cholera. In six organic flocks, outbreaks of blackhead were diagnosed. Concurrent infections of erysipelas and blackhead were diagnosed in one organic flock. Escherichia coli infections in the form of septicemia were identified in all organic flocks. In addition, cannibalism and constipation contributed significantly to the mortality in some organic flocks. In the confined deep litter flocks, E. coli infection, constipation, and cannibalism represented the most common causes of mortality.     

Risk factors for foot-pad dermatitis in chicken and turkey broilers in France

A prospective cohort study was conducted in the Brittany region to identify the risk factors related to foot-pad dermatitis (considered an indicator of animal well-being) in chicken and turkey broilers reared under commercial conditions. Factors related to the shed, equipment, litter management and stocking density were recorded; the dependent variable was the prevalence of lesions observed on the slaughterhouse chain. Lesions were scored from 0 (no lesion) to 3 (severe lesion). Our survey lasted from May 1999 to October 2000. Fifty flocks of chicken broilers (15 farms), 27 flocks of female turkey broilers (21 farms) and 41 flocks of male turkey broilers (27 farms) were surveyed. In chicken broilers, 10% of flocks were of high quality (80% of birds with score 0) and this was related to the use of concrete floors with a thin layer of wood shavings. In turkey broilers, 48% of female and 46% of male flocks were of bad quality (>10% of birds with score 3). A poor fan ventilation system (<150 m³/h/kg) was a significant risk factor. Turkey flocks of high quality were not observed. Stocking density had no influence on the prevalence of foot-pad dermatitis. We concluded that it is possible under high commercial stocking densities to have flocks with a low prevalence of foot-pad dermatitis in chicken broilers, whereas it is not in turkey broilers. Hence in chicken broilers, implementing a monitoring system based on the observation of foot-pad dermatitis prevalence at slaughter appears to be more appropriate than to legislate stocking density. In turkey broilers, it would probably be necessary either to reduce the stocking density drastically or to investigate new systems of floor drainage.     

Temperature-related risk factors associated with the colonization of broiler-chicken flocks with Campylobacter spp. in Iceland, 2001-2004

Our objective was to identify temperature-related risk factors associated with the colonization of broiler-chicken flocks with Campylobacter spp. in Iceland, with an underlying assumption that at minimum ambient temperatures, flies (Musca domestica) play a role in the epidemiology and seasonality of Campylobacter. At slaughter, pooled caecal samples were obtained from 792 flocks that hatched between March 15 and September 15 each year from 2001 to 2004, and cultured for Campylobacter. Daily temperature data (average, maximum and minimum temperatures) were obtained from the Icelandic Meteorological Office. We determined the closest weather station to each farm using ArcGIS 9. We then used logistic-regression models (with a random-effects term for farm) to examine associations between flock positivity and temperature-related predictors. The overall prevalence of Campylobacter was 27.4%, and the month with the highest prevalence (54.4%) was August. The final model included cumulative degree-days (CDD) above an average temperature of 4.4 degrees C and the presence of 1 or more days below a maximum temperature of 8.9 degrees C (threshold below which fly activity was expected to be substantially reduced and egg laying was not expected to occur) during the period 2-4 weeks (i.e. days 8-28) before slaughter. Below 79 CDD, the risk of flock Campylobacter colonization was generally low. Between 79 and 139 CDD, the risk increased gradually, but was lower for flocks raised during periods with 1 or more days <8.9 degrees C than flocks raised without days <8.9 degrees C. The risk increased sharply under conditions of high CDD (>139) and when the maximum temperature remained >8.9 degrees C during the period 2-4 weeks before slaughter.