PROCESSING AND PRODUCTS. Reactions of Laying Hens and Broilers to Different Gases Used for Stunning Poultry

Transcription

1 PROCESSING AND PRODUCTS Reactions of Laying Hens and Broilers to Different Gases Used for Stunning Poultry A. B. Webster 1 and D. L. Fletcher Department of Poultry Science, Poultry Science Building, The University of Georgia, Athens, Georgia ABSTRACT Observations of the behavioral reactions Chickens in 70% argon/30% CO 2 tended to demonstrate of laying hens and broilers to different gas stunning atmospheres were made. Sixty Hy-Line W-36 hens and 60 market-weight commercial broilers were placed individually into a plexiglass gas stunning unit and exposed to one of six gas atmospheres: air, concentrations of 30, 45, or 60% CO 2 in air, a mixture of 70% argon and 30% CO 2, and 100% argon. Video records were made during each test, which lasted until the subject became unconscious or for 2 min in the air treatment. in the 100% argon atmosphere resembled that in air, until birds became impaired by anoxia. All treatments involving CO 2, including 70% argon/30% CO 2, caused deep breathing and head shaking. The concentration of CO 2 in air in the range tested did not affect the tendency to perform different actions, except that birds in 60% CO 2 were more likely to exhibit a convulsive flip at the point of collapse. less sedation and performed more sudden efforts to regain balance during tests than did chickens in CO 2 mixtures in air and were more likely to perform a convulsive flip. Deep breathing and head shaking have been suggested as being indicative of respiratory distress and aversive reaction to CO 2. The data in this study are consistent with the possibility that head shaking is an alerting response functioning to promote arousal in the face of reduced sensibility during exposure to CO 2 -enriched atmospheres. Nonetheless, if the view is correct that deep breathing and head shaking indicate distress, the 70% argon/30% CO 2 gas mixture was at least as distressing as even 60% CO 2 in air. The relative prevalence of sudden efforts to regain balance in 70% argon/30% CO 2 suggest that this gas mixture might cause even more distress than up to 60% CO 2 in air. (Key words: gas stunning, laying hen, broiler, carbon dioxide, argon) 2001 Poultry Science 80: INTRODUCTION The idea to use gas to stun poultry prior to slaughter is not new (Kotula et al., 1961). In recent years, economic pressure to improve handling efficiency and carcass quality, as well as animal welfare concerns regarding shackling prior to electrical stunning, has renewed interest in gas stunning technology. CO 2, N, Ar, Ar/CO 2 mixtures, and mixtures of O 2,CO 2, and N are effective for gas stunning when used in appropriate concentrations (Mohan Raj and Gregory, 1990a; Mohan Raj et al., 1992a; Poole and Fletcher, 1995; Raj et al., 1998). Because CO 2 is involved in gas exchange with oxygen in the lungs, an increased atmospheric partial pressure of CO 2 interferes with blood oxygenation. This relationship creates two advantages for the use of CO 2 in air for gas stunning. Cost is reduced because less CO 2 is needed. Second, moderate concentrations of CO 2 in air are easier to maintain, allowing wider margins for error than for alternate stun Poultry Science Association, Inc. Received for publication February 7, Accepted for publication May 11, To whom correspondence should be addressed: bwebster@uga.edu. ning gas atmospheres, which must dilute air to a low residual level to be effective (Mohan Raj and Gregory, 1990b; Mohan Raj et al., 1992a). This second advantage is especially important for on-farm, modified-atmosphere killing situations in which rigid control of gas concentration may be difficult (Webster et al., 1996a,b). CO 2 has an anesthetic effect when inhaled (Andrews et al., 1993), which would reduce sensation of a bird during gas stunning. This effect gives a welfare advantage to the use of CO 2 in gas stunning situations because it would reduce distress, particularly of birds already in pain, e.g., due to leg disorder or injury. Some researchers, however, consider the use of high concentrations of CO 2 to stun poultry to be unacceptable on humanitarian grounds because the gas itself is likely, in their opinion, to be unpleasant or distressing to the birds (Mohan Raj et al., 1992 a,b; Raj and Gregory, 1993). They suggest that Ar or a mixture of Ar and CO 2 (to 30%) be used for gas stunning on the premise that these gasses are more likely to be humane than CO 2 in air. Hens learn to avoid areas with 7.5% CO 2 in air or where the O 2 was only 10%, achieved by diluting the air with Ar (Raj and Gregory, 1991). These aversions did not appear to be strong because they were over-ridden by social factors. 1371

2 1372 WEBSTER AND FLETCHER Raj (1994) and Raj et al. (1998) mention unpublished observations in which hens showed no aversion to an Ar atmosphere where food was placed, but three of eight hens would not enter a 48% CO 2 atmosphere to consume feed. The fact that five of eight hens did enter the CO 2 atmosphere in a free-choice situation suggests that aversion to the CO 2 was not great. There has been little study of the behavior of chickens upon entering atmospheres of air enriched with CO 2. Furthermore, insight into the relative welfare impacts of various CO 2 /air mixtures, Ar, and Ar/CO 2 mixtures might be gained by comparing the reactions of chickens in these different atmospheres. The present study, therefore, describes the actions of laying hens and chicken broilers in atmospheres of 30, 45, and 60% CO 2 in air, 70% Ar and 30% CO 2, and 100% Ar. MATERIALS AND METHODS The hens used in the study were a commercial layer variety (Hy-Line W-36), 62 or 101 wk of age, and were drawn from a flock housed in battery cages at the University of Georgia. The older hens had been molted. Broilers (approximately 6 wk old) were obtained from the live bird holding area of a local commercial processing plant and were held overnight with feed and water before being used in the stunning trials. The flow-through gas stunning unit has been described by Poole and Fletcher (1995). The unit comprised a plexiglass box ( cm) with 1.9 cm diameter inlet and exhaust ports on opposite ends of the box. Inside the apparatus, a plexiglass baffle plate ( cm) was mounted on the end wall in front of the inlet port leaving a 2.5 cm space between plate and wall. The baffle plate prevented incoming gases from blowing directly onto a test bird and helped to distribute the gas throughout the chamber. The chamber was used as described by Poole and Fletcher (1995), except that the birds were placed directly into the unit without first being restrained in a wire mesh cage. The six gas treatments in the study were air (control); concentrations of 30, 45, and 60% CO 2 in air (CO 2 /air treatments); a mixture of 70% argon and 30% CO 2 (Ar/ CO 2 ); and 100% argon (Ar). All of these gases were dispensed from pressurized gas cylinders through appropriate pressure regulators as described by Poole and Fletcher (1995). Ten hens and ten broilers were tested in each gas atmosphere (60 hens and 60 broilers total). Each bird was tested once only. In each trial, a single bird was placed into the gas stunning unit, facing the baffle plate on the inlet side. After the lid of the chamber had been replaced, the test gas was released into the chamber continuously until the bird became unconscious, or for 2 min when using air. The bird was then removed, and the chamber was prepared for the next trial. Video cameras were positioned on each of the long sides of the gas stunning box so as to acquire a clear view of a test bird from both sides. The simultaneous outputs of these cameras were delivered by means of a quad splitter to a time-lapse video cassette recorder that recorded the combined video output onto a videotape along with a continuous record of the time course of the trial. The following behavioral actions were recorded from the videotapes: 1. Mouth and Throat Movements. Actions of the lower mandible, tongue, and throat suggestive of tasting. These actions were unrelated to gular movements associated with thermoregulatory panting. 2. Normal Breaths. The number of normal breaths taken by a chicken in a stunning trial before deep breathing was initiated was counted. 3. Deep Breathing. Deepened breathing distinctly different from the normal breathing rhythm or from thermoregulatory panting, characterized by deeper than normal inspiration through the mouth, generally accompanied by extension of the neck. In atmospheres containing CO 2, this action, once started, would become successively more pronounced over the first few breaths. 4. Head Shake. A series of quick lateral rotational movements of the head. The behavioral pattern was performed with varying degrees of intensity. Head shaking in Ar was different from the normal action. Instead of being a behavioral pattern with clear beginning and end, as is true of a typical head shake, it was performed more or less continuously once started, continuing until the bird collapsed, and consisted of individual rotational jerks of the head from side to side. 5. Recovery of Balance. A sudden effort to regain balance accompanied by lifting of the wings away from the body. 6. Defecation. Voiding of feces. Defecation was recorded only for broilers because it occurred rarely among hens. 7. Stillness. An episode of stillness of the body and head, sometimes with gradual forward subsidence of the head toward the floor, suggestive of sedation. 8. Eye Closure. Sustained closure of the eyes. 9. Flip. A convulsive action immediately preceding total collapse that tended to flip the bird onto its back. 10. Loss of Posture. Depending on the gas in which the bird was tested, this behavior was identified by loss of tension in the neck while recumbent or by falling over with no apparent righting response. It was not unusual for chickens in 30% CO 2 in air to sit propped against the chamber wall where they could not fall over. With these, loss of neck tension was taken to indicate loss of posture. Birds having manifested loss of posture were unresponsive to touch and appeared to be totally unaware. Loss of posture is considered an indicator of unconsciousness (Raj et al., 1998). For each of the behaviors listed above, except normal breaths, the time when the behavior was first performed in a trial was recorded. In addition, the number of deep

3 REACTIONS OF CHICKENS TO STUNNING GASES 1373 TABLE 1. Percentage of laying hens in each treatment that showed the indicated behavioral actions during gas stunning trials Mouth Recovery and Deep Head of Eye Loss of Treatment throat breathing shake balance Stillness closure Flip posture Air % CO % CO % CO Ar/CO Ar Results of chi-squared comparisons CO 2 /air 1 vs. air ** ** ** * ** ** * ** Among % CO 2 NS NS NS NS NS NS * NS CO 2 /air vs. Ar/CO 2 NS NS NS NS ** ** * NS CO 2 /air vs. Ar ** ** NS NS ** NS * NS Ar/CO 2 vs. air ** ** ** ** * NS ** ** Ar/CO 2 vs. Ar ** ** NS * * ** NS NS Ar vs. air NS NS * NS NS ** * ** 1 30, 45, or 60% CO 2 in mixtures with air. *P < **P < breaths, head shakes, and recoveries of balance that occurred until loss of posture were counted. The data were analyzed separately for layers and broilers. Chi-squared contingency analyses were calculated to compare the number of birds in each treatment that manifested given actions. The times taken to initiate specific actions (latencies) were compared by analyses of variance by the using general linear models program of SAS software (SAS Institute, 1993). Tukey s studentized range tests were used to assess the significance of differences between treatment means. At least half of the birds in a treatment had to perform a given behavior for that treatment to be included in the analysis of variance pertaining to the behavior. RESULTS AND DISCUSSION Prevalence of As expected, there were few responses to the air treatment. Among the hens, two manifested mouth and throat movements, and four performed head shakes (Table 1). For broilers, the corresponding numbers of birds that performed these actions were nine and one, respectively (Table 2). in the Ar atmosphere was similar to that in air during the initial portion of the stunning test. The numbers of birds showing mouth and throat actions, deep breathing, defecation, and stillness did not differ in air vs. Ar for hens or broilers (Tables 1 and 2). Obvious differences in behavior between the air and Ar treatments did not appear until the birds became impaired in the Ar atmosphere. All the treatments containing CO 2 in air produced marked differences in behavior compared to the air treatment (Tables 1 and 2). Significantly more birds in CO 2 / air treatments showed mouth and throat movements (layers), deep breathing, head shake, recovery of balance, defecation (broilers), stillness, eye closure, flip, and loss of posture. For layers and broilers, tendencies to manifest the different actions were not affected by CO 2 concentrations in air ranging from 30 to 60%, except for flip, which for layers was significantly more likely to occur at 60% CO 2. in the Ar/CO 2 atmosphere closely resembled that in the CO 2 /air atmospheres. For layers, however, hens in the Ar/CO 2 tests were significantly less likely to show Stillness and Eye Closure and were more likely to perform flip. Broilers in Ar/CO 2 also had a higher incidence of Flip. The greater prevalence of stillness for layers in CO 2 /air atmospheres suggests a calming effect of these gas mixtures during the latter stages of the stunning process. Similarly, fewer broilers manifested Stillness in Ar/CO 2 than in the CO 2 /Air treatments, but the difference in numbers was not statistically significant. Flip was most prevalent in the most anoxic environments. For layers, the incidence of Flip did not differ among 60% CO 2, Ar/CO 2, and Ar (chi-squared values for 60% CO 2 vs. Ar/CO 2 and vs. Ar, respectively, were 3.2 and 0.2; both were not statistically significant at P > 0.05). All of the broilers in the Ar/CO 2 and Ar treatments manifested flip, whereas only half of the broilers in 60% CO 2 did so. This difference was statistically significant (chi-squared = 6.66; P < 0.01). al Latencies For layers and broilers, the order in which behavioral actions appeared, on average, during stunning in atmospheres containing CO 2 was mouth and throat movements, deep breathing, head shake, recovery of balance, defecation (broilers), stillness (when evident), eye closure,

4 1374 WEBSTER AND FLETCHER TABLE 2. Percentage of broilers in each treatment that showed the indicated behavioral actions during gas stunning trials Mouth and Deep Head Recovery Eye Loss of atment throat breathing shake of balance Defecation Stillness closure Flip posture Air % CO % CO % CO Ar/CO Ar Results of chi-squared comparisons CO 2 /air 1 vs. air NS ** ** ** ** ** ** * ** Among % CO 2 NS NS NS NS NS NS NS NS NS CO 2 /air vs. Ar/CO 2 NS NS NS NS NS NS NS ** NS CO 2 /air vs. Ar ** ** NS NS ** ** NS ** NS Ar/CO 2 vs. air NS ** ** ** ** ** ** ** ** Ar/CO 2 vs. Ar * ** NS NS ** ** NS NS NS Ar vs. air NS NS ** ** NS NS ** ** ** 1 30, 45, or 60% CO 2 in mixtures with air. *P < **P < and loss of posture (Tables 3 and 4). The longer time apparently taken to show eye closure than to exhibit loss of posture for layers under 45% CO 2 is an artifact. The mean latency for eye closure was calculated using only the hens that manifested the behavior. Between four and six breaths were taken after the start of gas injection into the chamber before initiation of deep breathing. For laying hens, head shaking was first performed an average of 6.4 s (30% CO 2 ), 5.2 s (45% CO 2 ), 3.1 s (60% CO 2 ), or 2.7 s (Ar/CO 2 ) after mouth and throat movements. For broilers, these times were 11.7 s (30% CO 2 ), 2.2 s (45% CO 2 ), 2.3 s (60% CO 2 ), and 10.0 s (Ar/CO 2 ). Head shaking in the Ar atmosphere began considerably later than it did in atmospheres containing CO 2, and, as described in the Materials and Methods section, it was performed differently. Tables 3 and 4 show results of the statistical comparisons of latencies to perform different actions in the various stunning atmospheres. Although an attempt was made to dispense all gases into the stunning chamber at the same rate, different gas regulators were used to fit the attachments specific to the different types of gas cylinder. Some of the differences in behavioral latency might have been due to variations in the rate at which the different gases filled the test chamber. For the mixtures of CO 2 in air, the times to loss of posture were considerably shorter in higher concentrations of CO 2, although there was little difference in latency to unconsciousness between 45% CO 2 and 60% CO 2. Time to loss of posture did not differ significantly among 60% CO 2, Ar/CO 2, and Ar, with average times ranging from 32 to 35 s for layers and 28 to 35 s for broilers. Onset of loss of posture was most gradual in 30% CO 2. Once deep breathing started, the number of deep breaths taken prior to loss of posture did not differ significantly between the higher concentrations of CO 2 in air and Ar/CO 2, averaging eight to nine breaths for layers and broilers (Table 5). Significantly more deep breaths were taken by birds in 30% CO 2 prior to loss of posture. Hens in Ar/CO 2 tended to perform the greatest number of recoveries of balance prior to loss of posture (Table 5). The same pattern was evident for broilers. TABLE 3. Latency to first performance of behavioral actions by laying hens in gas stunning trials 1 Mouth Normal Deep Head Recovery Eye Loss of and throat breaths breathing shake of balance Stillness closure posture Treatment (s) (n) (s) (s) (s) (s) (s) (s) Air % CO ab b 12.7 b a 60.5 a 61.5 a 45% CO b b 11.2 b b 40.2 b 35.8 b 60% CO ab b 9.2 b b 35.0 b 34.9 b Ar/CO a a 11.5 b b Ar a b 35.4 b SEM a,b Means that differ significantly within columns have no common superscripts (P 0.05). 1 Empty cells correspond to treatments in which too few hens showed the behavior to warrant inclusion in the analysis.

5 REACTIONS OF CHICKENS TO STUNNING GASES 1375 TABLE 4. Latency to first performance of behavioral actions by broilers in gas stunning trials 1 Mouth Normal Deep Head Recovery Eye Loss of and throat breaths breathing shake of balance Defecation Stillness closure posture Treatment (s) (n) (s) (s) (s) (s) (s) (s) (s) Air 22.8 a % CO b ab 15.9 b 24.0 a a 46.1 a 57.0 a 45% CO b ab 6.4 c 19.0 b ab 34.8 b 37.9 b 60% CO b b 6.7 c 15.4 b ab 32.7 b 35.4 bc Ar/CO b a 15.4 b 19.5 b b 28.3 bc 28.5 c Ar 9.3 ab a 28.2 a c 30.7 bc SEM a c Means that differ significantly within columns have no common superscripts (P 0.05). 1 Empty cells correspond to treatments in which too few broilers showed the behavior to warrant inclusion in the analysis. Aversiveness of the Different Stunning Gases Deep breathing and head shaking of chickens in response to CO 2 in air have been suggested to indicate respiratory distress due to a possible sense of breathlessness and pungency of the CO 2 (Gregory et al., 1990; Raj and Gregory, 1994). This argument is based largely on the fact that humans find high concentrations of CO 2 to cause a feeling of breathlessness and often to be pungent (Gregory et al., 1990). Other gases or gas mixtures, such as Ar or a 70% Ar and 30% CO 2 mixture, have been promoted for gas stunning on the premise that stunning concentrations of CO 2 are not sufficiently humane (Mohan Raj et al., 1992a,b; Raj and Gregory, 1993, 1994). Although it is possible that deep breathing induced by CO 2 is distressing and that head shaking is a reaction to the pungency of the gas, neither of these propositions has been verified in chickens. Assessments of behavior in tests designed, for example, to create potential motivational conflict might provide insight into the relative levels of aversion chickens experience to different gas mixtures. Because the present study did not test actual aversion to gases, it does not allow conclusions as to the welfare significance of CO 2 -induced deep breathing and head shaking. However, the order of appearance of different actions in the CO 2 -enriched atmospheres does permit an alternate suggestion for the cause of head shaking, albeit not tested in the current investigation. Chickens took several normal breaths as incoming gas displaced air in the chamber during stunning trials. The birds made their first response to CO 2 gas mixtures by manifesting mouth and throat movements. Shortly thereafter deep breathing began. Head shaking usually began after one or more deep breaths. Lambooij et al. (1999) observed a similar sequence of behavior in broilers. In that study, head shaking also did not occur immediately upon immersion in CO 2 -enriched gases but tended to appear after deep breathing had begun. Head shaking is a common action of chickens and has been suggested in other circumstances to be an alerting response functioning to promote arousal (Hughes, 1983). Along these lines, Levy (1944) stated that head shaking probably was at a maximum when hens were falling asleep, and Webster (2000) noted that frequency of head shaking was higher when hens were in a somnolent state than when they were alert. Following Hughes interpretation (Hughes, 1983), head shaking could be an attempt to regain alertness in response to reduced sensibility as CO 2 -induced anesthesia takes effect. If irritation of mucosal membranes was the cause of head shaking, one might expect that head shaking would be among the first actions to occur in a CO 2 -enriched atmosphere and that it would be vigorous at first, dimin- TABLE 5. Number of deep breaths, head shakes, and recoveries of balance in gas stunning trials 1 Laying hens Broilers Deep Head Recoveries Deep Head Recoveries Treatment breaths shakes of balance breaths shakes of balance Air % CO a 3.7 ab 1.2 ab 16.1 a b 45% CO b 2.8 b 0.8 ab 8.9 b b 60% CO b 3.6 ab 0.6 b 8.3 b b Ar/CO b 5.3 a 1.9 a 8.3 b a Ar ab SEM a,b Means that differ significantly within columns have no common superscripts (P 0.05). 1 Empty cells correspond to treatments in which too few birds manifested the behavior to warrant inclusion in the analysis.

6 1376 WEBSTER AND FLETCHER ishing in intensity as anesthesia developed. Alternatively, if head shaking arose from attempts to maintain alertness, it would occur only after the bird had inhaled enough CO 2 to affect neural function. In this scenario, head shaking would not diminish in strength and perhaps would become stronger during the stunning process. Although the strength of the head shaking response was not quantified in the present study, it was evident, empirically, that the first events of head shaking often were not as obvious or performed in such a fully elaborated fashion as those later. Head shaking did not appear to diminish in strength as stunning progressed. In another experiment, broilers entering CO 2 -enriched atmospheres tended to manifest head shaking before deep breathing (Gerritzen et al., 2000). There is no evident reason for this apparent reversal of sequence. Head shaking is a normal behavior in situations unrelated to gas stunning. Causal stimuli for head shaking might have been present in the study by Gerritzen et al. (2000) that were not part of the present study or of the work by Lambooij et al. (1999). Assuming that a given behavior has the same welfare significance regardless of the gas atmosphere in which it is performed, the data of the present study allow some comparison of the welfare impacts of the gas mixtures tested. Mohan Raj et al. (1992a,b) and Raj and Gregory (1993) contend that a mixture of 30% CO 2 in Ar is more humane for the gas stunning of poultry than concentrations of CO 2 in air sufficient to dispatch birds quickly. In the present study, however, every action that was performed in trials using CO 2 /air mixtures was also performed in the trials with the Ar/CO 2 gas mixture. Lambooij et al. (1999) and Gerritzen et al. (2000) have also reported occurrence of deep breathing and head shaking in Ar/CO 2. If it is supposed that deep breathing and head shaking indicate distress, then Ar/CO 2 apparently would have as much negative welfare impact as even 60% CO 2 in air. In fact, Ar/CO 2 might even be worse. Test birds in the CO 2 /air atmospheres usually appeared to become gradually sedated and less responsive to their surroundings as the tests progressed. This result was less true in Ar/CO 2 tests wherein chickens tended to show more recoveries of balance than did birds in the 45% and 60% CO 2 /air atmospheres. Chickens in Ar/CO 2 gave a greater appearance of sustained struggle to maintain balance prior to loss of posture. Reduction of ability to maintain balance while a chicken still has high awareness may be distressing to the bird. All of the gases tested would be effective for gas stunning. Their actual efficacy would depend on requirements for time to stun and carcass quality. A quick stun can be achieved using moderate to high concentrations of CO 2 in air or high concentrations of pure Ar or Ar/CO 2 mixtures. If carcass damage is an issue, it might be better to use 30% CO 2 in air. For example, we have observed in our laboratory that convulsions are reduced if death occurs after a bird is rendered unconscious in a 30% CO 2 / air atmosphere, whereas convulsions can be very vigorous after stunning in Ar, Ar/CO 2, or a mixture of 60% CO 2 /air (unpublished data). ACKNOWLEDGMENTS This study was supported by a grant from the U.S. Poultry and Egg Association and by state and HATCH funds allocated to the University of Georgia Agricultural Experiment Station, Athens, GA REFERENCES Andrews, E. J., B. T. Bennett, J. D. Clark, K. A. Houpt, P. J. Pascoe, G. W. Robinson, and J. R. Boyce, report of the AVMA panel on euthanasia. J. Am. Vet. Med. Assoc. 202: Gerritzen, M. A., E. Lambooij, S. J. W. Hillebrand, J. A. C. Lankhaar, and C. Pieterse, al responses of broilers to different gaseous atmospheres. Poultry Sci. 79: Gregory, N. G., A. B. Mohan Raj, A. R. S. Audsey, C. C. Daly, Langford, Effects of CO 2 on man. Fleischwirtsch 70: Hughes, B. O., Headshaking in fowls: The effect of environmental stimuli. Appl. Anim. Ethol. 11: Kotula, A. W., E. E. Drewniak, and L. L. Davis, Experimentation with in-line carbon dioxide immobilization of chickens prior to slaughter. Poultry Sci. 40: Lambooij, E., M. A. Gerritzen, B. Engel, S. J. W. Hillebrand, J. Lankhaar, and C. Pieterse, Behavioural responses during exposure of broiler chickens to different gas mixtures. Appl. Anim. Behav. Sci. 62: Levy, D. M., On the problem of movement restraint: Tics, stereotyped movements, hyperactivity. Am. J. Orthopsychiatry 14: Mohan Raj, A. B., and N. G. Gregory, 1990a. Effect of rate of induction of carbon dioxide anaesthesia on the time of onset of unconsciousness and convulsions. Res. Vet. Sci. 49: Mohan Raj, A. B., and N. G. Gregory, 1990b. Investigation into the batch stunning/killing of chickens using carbon dioxide or argon-induced hypoxia. Res. Vet. Sci. 49: Mohan Raj, A. B., and N. G. Gregory, Preferential feeding behaviour of hens in different gaseous atmospheres. Br. Poult. Sci. 32: Mohan Raj, A. B., N. G. Gregory, and L. J. Wilkins, 1992a. Survival rate and carcase downgrading after the stunning of broilers with carbon dioxide-argon mixtures. Vet. Rec. 130: Mohan Raj, A. B., S. B. Wotton, and N. G. Gregory, 1992b. Changes in the somatosensory evoked potentials and spontaneous electroencephalogram of hens during stunning with a carbon dioxide and argon mixture. Br. Vet. J. 148: Mohan Raj, A. B., S. B. Wotton, and P. E. Whittington, 1992c. Changes in the spontaneous and evoked electrical activity in the brain of hens during stunning with 30 percent carbon dioxide in argon with 5 per cent residual oxygen. Res. Vet. Sci. 53: Poole, G. H., and D. L. Fletcher, A comparison of argon, carbon dioxide, and nitrogen in a broiler killing system. Poultry Sci. 74: Raj, A. B. M., An investigation into the batch killing of turkeys in their transport containers using mixtures of gases. Res. Vet. Sci. 56: Raj, A. B. M., S. B. Wotton, J. L. McKinstry, S. J. W. Hillebrand, and C. Pieterse, Changes in the somatosensory evoked potentials and spontaneous electroencephalogram of broiler chickens during exposure to gas mixtures. Br. Poult. Sci. 39: Raj, M., and N. G. Gregory, Time to loss of somatosensory evoked potentials and onset of changes in the spontaneous

7 REACTIONS OF CHICKENS TO STUNNING GASES 1377 electroencephalogram of turkeys during gas stunning. Vet. Rec. 133: Raj, M., and N. G. Gregory, An evaluation of humane gas stunning methods for turkeys. Vet. Rec. 135: SAS Institute, Statistical Analysis System proprietary software release SAS Institute Inc., Cary, NC. Webster, A. B., of White Leghorn laying hens after withdrawal of feed. Poultry Sci. 79: Webster, A. B., D. L. Fletcher, and S. I. Savage, 1996a. Humane on-farm killing of spent hens. J. Appl. Poult. Res. 5: Webster, A. B., D. L. Fletcher, and S. I. Savage, 1996b. Handling spent hens on the farm. Pages in: Proceedings of the 1996 National Poultry Waste Management Symposium. P. H. Patterson and J. P. Blake, ed. Publisher: the National Poultry Waste Management Symposium Committee. Auburn University Printing Service, Auburn, AL.