Humane slaughter for fish Vis_et_al-2003-Aquaculture_Research

SAVE OFFLINE

Aquaculture Research, 2003, 34, 211±220

Is humane slaughter of fish possible for industry? Hans van de Vis1, Steve Kestin2, David Robb2, JoÈrg OehlenschlaÈger3, Bert Lambooij4, Werner MuÈnkner3, Holmer Kuhlmann5, Karin Kloosterboer1, Margarita Tejada6, Almudena Huidobro6, HaÊkon OtteraÊ7, Bjùrn Roth8, Nils Kristian Sùrensen9, Leif Akse9, Hazel Byrne10 & Paul Nesvadba10 1

The Netherlands Institute for Fisheries Research (RIVO), IJmuiden, the Netherlands

2

University of Bristol, Department of Clinical Veterinary Science, Bristol, UK

3

Federal Research Centre for Fisheries, Institute for Fishery Technology and Fish Quality, Hamburg, Germany Institute for Animal Science and Health (ID-Lelystad), the Netherlands

4 5

Federal Research Centre for Fisheries, Institute for Fishery Ecology, Hamburg, Germany

6

Instituto del Frio (CSIC), Ciudad Universitaria s/n, Madrid, Spain

7

Institute of Marine Research, Department of Aquaculture, Bergen, Norway

8

University of Bergen, Institute for Fisheries and Marine Biology, Bergen, Norway

9

Norwegian Institute of Fisheries and Agriculture, Tromsù, Norway

10

Robert Gordon University, Food Science and Technology Research Centre, Aberdeen, UK

Correspondence: Hans van de Vis, The Netherlands Institute for Fisheries Research (RIVO), PO Box 68, 1970 AB Ijmuiden, The Netherlands. E-mail: [email protected]

Abstract The objective was to evaluate industrial and research slaughter methods for Atlantic salmon (Salmo salar), gilt-head seabream (Sparus auratus) and eel (Anguilla anguilla) with respect to welfare and quality. As a general term of reference, an optimal slaughter method should render fish unconscious until death without avoidable excitement, pain or suffering prior to killing. For Atlantic salmon, commercial slaughter methods (carbon dioxide stunning followed by gill cutting, and gill cutting alone) are not in conformity with the general term of reference, as the fish are not rendered unconscious immediately and possibly experience stress. Evaluation of automated percussive stunning remained unconclusive. More research should enable us to ascertain whether loss of consciousness is instantaneous. Electrical stunning can be humane if applied properly. However, because flesh of electrostunned fish was characterized by occasional bloodspots, optimization of the electrical parameters is required. Prototypes for percussive and electrical stunning of salmon have been recently developed. This implies that humane slaughter of salmon is feasible for industry. For gilt-head seabream, neither aphyxia in air nor transfer of the fish to an ice ß 2003 Blackwell Publishing Ltd

slurry were considered to be humane: the methods did not induce immediate brain dysfunction and vigorous attempts to escape occurred. Percussive and electrical stunning can be in conformity with the general term of reference. However, conditions for stunning whole batches of seabream have not been established. Quality of the fish slaughtered by percussive stunning was similar to that obtained by the industrial method, i.e. immersion in an ice slurry. Further work is required to establish optimal stunning conditions and to develop prototypes. For eel, desliming in a salt-bath followed by evisceration, electrical stunning performed under the conditions prescribed by the German legislation, and live chilling and freezing were not considered to be humane. In contrast, it was established that a 10±20 kg batch of eels in fresh water could be rendered unconscious immediately and until death by applying electricity in combination with nitrogen gas. The conditions used were 0.64 A dm 2 for 1 s, followed by 0.17 A dm 2 combined with nitrogen flushing for 5 min. A preliminary assessment of flesh quality suggests that it may be improved by application of the latter method, compared with the salt bath. The results clearly indicated that humane slaughter of eels is possible in practice. 211

Is humane slaughter of fish possible for industry? Hans van de Vis et al.

Aquaculture Research, 2003, 34, 211±220

Keywords: Atlantic salmon, eel, fish flesh quality, gilt-head seabream, humane slaughter, stunning

applicability of humane slaughter in the fish farming industry.

Introduction

Methodology for welfare assessments

Food quality is nowadays perceived in the EU as a global concept. Food should be safe, tasty and healthy. Ethical aspects concerning food production are also of importance, such as protection of the environment and animal welfare. In this context, retailers (Cooke 2001), animal welfare associations and governments (TierSchlV 1997, 1999) are increasingly demanding humane methods to convert live farmed fish into food. In addition, the EU Directive on the protection of animals at slaughter (Anonymous 1993) applies to all vertebrates and states that `Animals shall be spared any avoidable excitement, pain or suffering during movement, lairaging, restraint, stunning, slaughter or killing'. Comparison of results from research carried out on fish, mammals and birds reveals that they present similarities in anatomy (neuro)physiology and behaviour (Chervova 1996; Mathews & Wickelgren 1978; Overmier & Hollis 1990; Verheijen & Flight 1997; Wendelaar Bonga 1997; Wiepkema 1997; Spruijt 1999). As a consequence, the requirements for humane slaughter of warm-blooded animals can be used to propose the following general term of reference to protect fish at slaughter: the animals should be rendered unconscious until death without avoidable excitement, pain or suffering. Current slaughter methods for farmed fish (e.g. death in air, death in ice slurry, carbon dioxide narcosis in combination with exsanguination) are in most cases not in accordance with the general term of reference (Robb 2001). Furthermore, it is likely that humane slaughter procedures could improve post mortem quality of fish, as reported for warm-blooded animals by many authors (Brown, Warriss, Nute, Edwards & Knowles 1998; Geesink, Mareko, Morton & Bickerstaffe 2001). Therefore, an EU collaborative project focused on the effects that various slaughter methods have on the welfare and quality of fish. Three commonly farmed species were used as models, namely Atlantic salmon (Salmo salar), gilt-head seabream (Sparus aurata) and eel (Anguilla anguilla). Research institutes and companies from Norway, the United Kingdom, Germany, Spain and The Netherlands collaborated. The present article presents selected findings of the project with emphasis on the

In order to establish whether commercial or research slaughter methods met the general term of reference, methodology to evaluate the degree of consciousness in fish was developed. A standardized protocol to monitor fish behaviour during slaughter was established (Kestin, Robb & Van de Vis 2002). The monitoring of behaviour was validated and applied to the three fish species. However, observation of behaviour only may not be sufficient for assessment of unconsciousness. For example, ineffective electrical stunning can be very painful and paralysis may occur without loss of consciousness (Croft 1952). Similarly, in the case of live chilling of eels, it appeared that the animals could become motionless without loss of consciousness (Lambooij, Van de Vis, Kloosterboer & Pieterse 2002a). Therefore, registration of evoked responses on electroencephlogram (EEG) recordings is recommended (Hoenderken 1978). Measurement of visual evoked responses (VERs) was performed on Atlantic salmon (Robb, Wotton, McKinstry, Sùrensen & Kestin 2000), gilt-head seabream (Robb & Kestin 2002) and eel (Lambooij, Van de Vis, Kloosterboer & Pieterse 2002b; Lambooij, Van de Vis, Kuhlmann, MuÈnkner, OehlenschlaÈger, Kloosterboer & Pieterse 2002c). A VER is the response in the brain to flashes of light directed towards the eyes. The absence of an average VER indicates brain dysfunction (Kestin, Wotton & Gregory 1991). In addition, the registration of somatosensory evoked responses (SERs), i.e. the responses in the brain to pain stimuli, was measured in eel (Lambooij et al. 2002a, b, c). The pain stimulus consisted of scratching the tail by using a needle. Typical changes in the pattern on the EEG, such as a general epileptiform insult, indicate unconsciousness. The absence of SERs on the EEG are indicative of insensibility.

212

Evaluation of slaughter methods with respect to welfare Atlantic salmon Atlantic salmon (2±3 kg) were raised in net cages in Tromsù (Norway) or in Loch Fyne (Scotland). Fish

ß 2003 Blackwell Publishing Ltd, Aquaculture Research, 34, 211±220

Aquaculture Research, 2003, 34, 211±220

Is humane slaughter of fish possible for industry? Hans van de Vis et al.

harvested in Loch Fyne were allowed to recover for 3 h prior to the experiments. Gill cut Exsanguination through a gill cut without stunning is a commercial slaughter method used in the UK and Norway (Robb et al. 2000). All four gill arches on one side of the head are severed. Death ensued from anoxia is brought about by ischaemia. It was observed that gill cutting is not in conformity with the general term of reference, as the VERs were not lost immediately and vigorous movements occurred (Robb et al. 2000).

However, in practice, the stun is often not immediate and fish are hit more than once (Wall 2001). For this reason a pneumatic gun, developed for humane stunning of small warm-blooded animals (Hewitt 1999), was used in the experiments to deliver the blow. This ensured both the accuracy and velocity of the impact (Robb et al. 2000). After percussive stunning, the ability to perform self-initiated behaviour was lost immediately, while loss of consciousness occurred after on average 0.3 min (Table 1). It appeared that the array of EEG electrodes hindered a correct application of the blow, which may have prevented immediate loss of brain function.

Carbon dioxide

Use of a hollow-punch

Carbon dioxide stunning followed by exsanguination through a gill cut is a method that is widely used commercially. It is recommended to stop the diffusion of gas when the pH of water is constant at 5.0, indicating that the water is saturated (Anonymous 1995). Fish are immersed in the water for a minimum of four minutes, subsequently removed from the tank and their gills cut. The data in Table 1 show that carbon dioxide stunning is not humane. It was observed that the fish were not rendered unconscious immediately and the animals moved vigorously during the application (Robb et al. 2000).

For the purposes of this trial the pneumatic stun gun was modified to fire a hollow bolt, 5 mm in diameter, to penetrate 29 mm into the head of the fish (Robb et al. 2000). Results in Table 1 show that brain function was not lost immediately after application of the method. Moreover, use of a hollow-punch does not appear feasible for industry, as a high level of precision is required due to the small size of fish brains. Precise application is also hindered by escape attempts if restraining is unsufficient (Robb & Kestin 2002).

Percussive stunning Percussive stunning is a widely used commercial method where the fish are hit on the head with a rapidly moving, manually applied club. Concussion is caused by the acceleration of the brain within the skull disrupting its function. When sufficient force is applied the concussion can be irrecoverable.

Whole body electrical stunning Electrical stunning using 50 Hz a.c., 4.6 A for 3 s in seawater was performed to establish whether the stun could be immediate (unpublished data). VERs and self-initiated behaviour were lost immediately (Table 1). The time to recovery of VERs was on average 4.8 min. Electrical stunning in the described conditions can be humane, provided that the fish are killed prior to recovery.

Table 1 Effect of various slaughter and stunning methods on loss of self-initiated behaviour and loss of brain function for Atlantic salmon

Gill cut Carbon dioxide Instrumental percussion Hollow punch Electricity (4.6 A for 3 s in seawater)

Average time to loss of self-initiated behaviour (min)

Average time to loss of VERs (min)

2.2 1.6 0 0 0

4.7 6.1 0.3 0.5 0

ß 2003 Blackwell Publishing Ltd, Aquaculture Research, 34, 211±220

213

Is humane slaughter of fish possible for industry? Hans van de Vis et al.

Gilt-head seabream Immature gilt-head seabream were raised in Cadiz, Spain. Fish with an average live weight of 262 g were fasted for 2 days and crowded with a net which was towed for 2 h on average. The animals were finally caught by a dipnet.

Aquaculture Research, 2003, 34, 211±220

Percussive stunning applied to the top of the head appeared to give the same results as the lateral application. Self-initiated behaviour was lost immediately and irrecoverably after application of the method. This suggests that percussive stunning can be a humane method. Head-only electrical stunning

Immersion in an ice slurry Ice was mixed with seawater in a 3:1 ratio. The temperature of the slurry was maintained at 0.8 + 0.2 8C during the process by adding ice if necessary. Immersion in an ice slurry is the current method used for gilt-head seabream (Huidobro, Pastor & Tejada 2000a). Ability to perform selfinitiated behaviour and VERs were lost after 5 min on average (Table 2). Furthermore, vigorous movements which may indicate stress were observed. Therefore the general term of reference was not met for this method. Apshyxia The former commercial method for farmed seabream consists of allowing the fish to die by asphyxia in air. The current slaughter method is immersion in an ice slurry followed by storage in ice. Measurement of brain function and monitoring of behaviour showed that neither removal of the fish to air, nor transfer of the fish to an ice slurry at 0.5 8C from water at 23 8C resulted in immediate brain dysfunction (Table 2). Moreover, attempts to escape were observed for both methods. The results clearly show that both methods are not in accordance with the general term of reference. Percussive stunning Percussive stunning was carried out using the previously described pneumatic gun. The head of the fish was hit laterally. The gun was operated at 6 bar.

For head-only stunning of gilt-head seabream, electrodes were placed laterally on the head to deliver the current (50Hz a.c., 80 V). In these conditions, a 1-s stun caused brain dysfunction in only one fish out of 10. This fish showed recovery of VERs 37 s after stunning. In contrast, the application of current across the head for 10 s stunned nine of the 10 fish. Of the nine fish stunned, three recovered VERs within 16 s, but the six remaining fish showed no recovery within 10 min, leading to the conclusion that they had been killed. Examination of the currents showed great variations. In the 1 s group, the current ranged from 27 mA to 200 mA, except for the one fish which was stunned which received 450 mA. In the 10 s duration group, the current rose above 400 mA for all fish. Seabream appear to require more than 200 mA across the head to be stunned. The exact minimum current to achieve stunning in all fishes remains to be determined.

Eel Prior to stunning, farmed eels were fasted in water at approximately 13 8C for at least 7 days. Salt bath In the Netherlands, the most commonly used slaughter method is the so-called salt bath. For desliming, NaCl or a combination of NaCl and aqueous Na2CO3 is added to live eels in a dry tank. The duration of the desliming process is on average

Table 2 Effect of various slaughter and stunning methods on loss of self-initiated behaviour and loss of brain function for gilt-head seabream

Asphyxia Immersion in an ice slurry Electricity Instrumental lateral percussion

214

Average time to loss of self-initiated behaviour (min)

Average time to loss of VERs (min)

4 5 0 0

5.5 5 0 ±

ß 2003 Blackwell Publishing Ltd, Aquaculture Research, 34, 211±220

Aquaculture Research, 2003, 34, 211±220

Is humane slaughter of fish possible for industry? Hans van de Vis et al.

20 min. Eels are subsequently eviscerated (Van de Vis, OehlenschlaÈger, Kuhlmann, MuÈnkner, Robb & Schelvis-Smit 2001). The method is not accordance with the general term of reference, as the VERs were not lost immediately (Table 3). Moreover, vigorous attempts to escape occurred for at least 3 min after addition of salt onto the animals (Van de Vis et al. 2001). Live chilling and freezing The procedure, consisting of live chilling in ice slurry followed by freezing in a cold brine, was assessed within the scope of a Dutch national project (Lambooij et al. 2002a). When the body temperature of the eels was below 5 8C they were transferred to a cold brine at 20 to 16 8C (Logtenberg 2000). Results presented in Table 3 show that live chilling did not meet the general term of reference, as vigorous attempts to escape occurred for 3 min and the electrocardiogram (ECG) revealed that the heart activity was irregular, which may indicate stress. Moreover, at least 5% of the eels were not stunned at a body temperature below 5 8C. Freezing in a cold brine to stun and kill eels was not humane either, since it took on average 27 s before a chilled eel, with a body temperature lower than 5 8C, was stunned by freezing of the brain. Bulk electrical stunning Since April 1999, the only method permitted for commercial slaughter of eel in Germany is electrical stunning. The recommended current is 0.1± 0.19 A dm 2 electrode surface (TierSchlV 1997, 1999). These conditions were obtained when 50 V

was applied, using fresh water of 500 mS and plate electrodes with a surface of 2450 cm2 (Kuhlmann & MuÈnkner 1996; Lambooij et al. 2002c). When such electrical conditions are applied for 1 s, the method is not humane, as it was observed that the VERs were not lost immediately (unpublished data). SERs could be detected on the EEG in four out of seven eels. A modification of the method, using a 17 A current (200 V, 50 Hz a.c.) for 1 s, was therefore evaluated (Lambooij et al. 2002c). Such conditions ensured that the eels were stunned, as a general epileptiform insult was provoked immediately (Table 3), which is indicative for unconsciousness. The self initiated behaviour was lost immediately after the application of the current. Decapitation Nowadays the method is applied rarely in The Netherlands. The method is not suitable for slaughter in bulk and is therefore applied only by retailers. The head of the animal is separated from its body by a cut behind the gills. The results in Table 3 show that decapitation of eels is not humane, as it was observed that it took on average 13 min before the VERs were lost. Electrical stunning in combination with flushing water with nitrogen gas For industry, an irrecoverable stun is required, as the batch-wise desliming and gutting may take 40±60 min. Lambooij et al. (2002c) established that a head-only stun with 250 V for 1 s in combination with a whole body stun with 50 V for 5 min could prolong the period of unconsciousness.

Table 3 Effect of various slaughter and stunning methods on loss of self-initiated behaviour and loss of brain function for eel

Salt bath Immersion in an ice slurry Immersion in a cold brine at 20 to Electricity 50 V for 5 min Electricity 50 V for 1 s Decapitation Electricity 200 V for 1 s Electricity 200 V for 1 s followed by 50 V ‡ nitrogen for 5 min

16 8C

Average time to loss of self-initiated behaviour (min)

Average time to loss of VERs (min)

Average time to loss of SERs (min)

± >3 ± > 0.5 ± ± 0

> 10 ± ± >5 ± 13 ±

± > 12 > 0.5 ± * ± 0²

0

±

0²

*A general epileptiform insult was induced in four out of seven fish. ²A general epileptiform insult was induced immediately in all fish.

ß 2003 Blackwell Publishing Ltd, Aquaculture Research, 34, 211±220

215

Is humane slaughter of fish possible for industry? Hans van de Vis et al.

Aquaculture Research, 2003, 34, 211±220

However, the animals could recover. This may be because of the ability of eels to exchange considerable quantities of oxygen and carbon dioxide across the skin (Tesch 1999), which may explain that the lack of respiratory movements during the stun is not sufficient to make it irrecoverable. Therefore, it was decided to investigate whether the eels could be stunned and killed prior to slaughter by using electricity in combination with a low oxygen level in the water. It was established that the eels could be stunned irrecoverably in fresh water by applying 200 V, 17 A for 1 s followed by 50 V, 4.2 A 50 Hz a.c. in combination with flushing the water for 5 min with nitrogen gas (International Patent WO 01/95732). During that period, the water oxygen content was reduced from 74 + 10 to 23 + 11% (Lambooij et al. 2002c).

No difference was found as regards fibrousness and chewiness of cooked flesh (data not shown). This is in contrast to findings from Roth, Veland, Moeller, Imsland & Slinde (2002a) and Sigholt, Erikson, Rustad, Johanen, Nordvedt & Seland (1997), who demonstrated that a firmer texture of salmon was associated with percussive stunning and low preslaughter stress. Automated percussive stunning was also compared with death by asphyxia. The decline in fish freshness over four days, assessed by the Torrymeter (MAFF 1989), is depicted in Fig. 2. There was a significant difference (P < 0.05) in the rate of decline in fish freshness between the two killing methods. The difference between the two groups became significant 26 h after death. The batch stunned by percussion remained fresher longer than the batch killed by asphyxia.

Evaluation of slaughter methods with respect to quality

Electrical stunning in seawater Electrical stunning of Atlantic salmon in a tank was performed on a large scale in a commercial slaughterhouse using frequencies between 500 and

Slaughter methods that could be humane were evaluated with respect to quality.

Atlantic salmon

10

Carbon dioxide Percussion

8

Score

Percussive stunning Automated percussive stunning was compared with carbon dioxide stunning in combination with a gill cut. The data presented in Fig. 1 show that no significant differences between two slaughter methods could be detected by sensory assessment of flavour of the cooked flesh. Similar results were obtained in previous experiments (Sùrensen 1999).

6 4 2 0 0

5

10

15

20

25

Days in ice

Figure 1 Influence of two slaughter methods on flavour of cooked Atlantic salmon fillet.

13

Torrymeter reading

12

Percussion

11

Asphyxia

10

9

8 0

10

20

30

40

50

60

Time (hours harvest)

216

70

80

90

100

Figure 2 Fish freshness of Atlantic salmon.

ß 2003 Blackwell Publishing Ltd, Aquaculture Research, 34, 211±220

Aquaculture Research, 2003, 34, 211±220

Is humane slaughter of fish possible for industry? Hans van de Vis et al.

1500 Hz, with stun durations between 6 and 12 s. No difference in texture parameters could be observed between the fish killed after electrical stunning and after carbon dioxide stunning. Nevertheless, bloodspots occurred specifically in the electrostunned fish (Fig. 3). Occurence of injuries was dependent on the voltage, stun duration and frequency (Roth, Imsland & Moeller 2002b). Robb, O'Callaghan, Lines & Kestin (2002) showed that it is possible to reduce this type of downgrading in rainbow trout by optimizing the electrical conditions.

the Texture Analyser TA-XT2i. A significant difference (P < 0.05) in hardness between the two batches was observed (see Fig. 5). Fillets obtained by the application of the humane method were harder than those from the salt bath. Visual inspection of the fillets did not reveal any bruises or bloodspots. These preliminary results suggest that the quality of the flesh may be improved. Morzel & Van de Vis (2002) described in more details the relationship between slaughter and product quality of eel.

Gilt-head seabream Implementation of humane slaughter in the fish farming industry Atlantic salmon The salmon industry has applied live chilling prior to stunning or killing for a few years. Live chilling limits the contamination of the carcasses compared with chilling in a gutted condition. However, Skjervold & Fjñra (2001) reported higher levels of stress indicators (plasma cortisol, glucose and lactate) in live-chilled salmon. Roth et al. (2002a) also

Ice slurry Percussion

16

QIM score

The relationship between sensory properties of whole seabream during storage and the slaughter method used was assessed, using the Quality Index Method (QIM). QIM is a rapid and objective scheme for grading of raw fish (Bremner 1985). Huidobro et al. (2000a) and Huidobro, Pastor, LoÂpez-Caballero & Tejada (2000b) adapted QIM for gilt-head seabream. In Fig. 4, it is shown that the QIM scores assigned to the batch slaughtered by percussion and gutting were similar to the ones assigned to the batch obtained by using chilling in ice slurry followed by gutting. An improvement of flesh quality had been expected with the use of percussive stunning. However, it is likely that any positive effects of humane slaughter was masked by the highly stressful preslaughter procedure (crowding for 2 h).

12 8 4 0 0

Eel

10

15

20

25

30

Days in ice

Figure 4 QIM scores of whole gilt-head seabream slaughtered by immersion in an ice slurry followed by gutting and percussion in combination with gutting.

3000 Hardness (g)

Eels slaughtered by electricity in combination with nitrogen gas were compared to eel slaughtered according to the salt-bath procedure. Eels from the two batches were brined, hot smoked, filleted and deskinned. Hardness of the fillets was measured by

5

2000

1000

0

Electricity + N2

Salt bath

Slaughter method

Figure 3 Bloodspots in Atlantic salmon after electrical stunning.

Figure 5 Effect of the application of a humane method and of the Dutch industrial method on hardness of hot smoked eel fillets.

ß 2003 Blackwell Publishing Ltd, Aquaculture Research, 34, 211±220

217

Is humane slaughter of fish possible for industry? Hans van de Vis et al.

showed that live-chilled fish had an early onset and resolution of rigor mortis. Preslaughter treatment was not the object of the present study but should be taken into account in any future work. Evaluation of various slaughter methods revealed that electrical stunning is a humane method, provided that the correct conditions are used (Roth et al. 2002b). It was however, observed that product quality could be affected due to the presence of bloodspots. Further work to optimize the conditions and prevent this defect is required. Similarly, more work is needed to establish with certainty whether automated percussive stunning renders salmon unconscious immediately. Nevertheless, prototype equipment for percussive and electrical stunning of Atlantic salmon have recently become available. A few processors of salmon in the UK and Norway are performing trials with the pieces of equipment. The prototype for electrical stunning is very efficient for stunning large amounts of fish. The economical feasibility of such equipment therefore appears promising.

Gilt-head bream The results obtained from measurement of brain function and observation of behaviour showed that percussive and electrical stunning can be humane. Percussive stunning resulted in a flesh quality similar to that obatined after immersion in ice slurry. At present, farmed seabream are slaughtered batch-wise and any alternative humane method should allow to slaughter fish in bulk. Further research is needed to establish the optimal conditions with respect to effective stunning and product quality, and to develop specific prototypes.

Eel For humane slaughter it is necessary that the eels are stunned immediately without recovery. It was observed that this is possible by the application of electricity in combination with nitrogen gas on the animals in a tank. Under these conditions, 10±20 kg batches of eels can be stunned (in Lambooij et al. 2002c). Furthermore, preliminary data on quality of fish flesh suggest that it could be improved by the application of the latter method. As a

218

Aquaculture Research, 2003, 34, 211±220

consequence, humane slaughter of eels appears feasible for industry.

Acknowledgments Martine Morzel is acknowledged for assistance in revision of the manuscript. The study was financed by the EU, contract FAIR CT97-3127. Additional funding for electrical studies in gilthead seabream was obtained from Spanish National funds (Project ALI-99±1262-CE). The Dutch Ministry of Agriculture, Nature Management and Fisheries sponsored the work on live chilling and freezing of eels.

References Anonymous (1993) On the protection of animals at the time of slaughter or killing. Council Directive 93/119/ EC. Official Journal of the European Communities no. 1340/ 21. Anonymous (1995) Operating Manual for the Product Certification Schemes for Scottish Quality Farmed Salmon and Smoked Scottish Quality Salmon. Scottish Quality Salmon Ltd, Inverness, Scotland. Bremner H.A. (1985) A convenient easy to use system for estimating the quality of chilled seafood. Fish Processing Bulletin 7, 59±70. Brown S.N., Warriss P.D., Nute. G.R., Edwards J.E. & Knowles T.G. (1998) Meat quality in pigs subjected to minimal preslaughter stress. Meat Science 49, 257±265. Chervova L.S. (1996) Pain sensitivity and behaviour of fishes. Journal of Ichthyology 37, 98±102. Cooke M. (2001) Ethical considerations for the production of farmed fish-the retailer's viewpoint. In: Farmed Fish Quality (ed. by S.C. Kestin. & P.D. Warriss), pp. 116±119. Blackwell, Oxford. Croft P.G. (1952) The effect of electrical stimulation of the brain on the perception of pain. Journal of Mental Science 42, 421±426. Geesink G.H., Mareko M.H.D., Morton J.D. & Bickerstaffe R. (2001) Effects of stress and high voltage electrical stimulation on tenderness of lamb m. longissumus. Meat Science 57, 265±271. Hewitt L. (1999) A novel stunning system for slaughter of poultry. In: Poultry Meat Science (ed. by R.I. Richardson & G.C. Mead), p. 422. CABI Publishing,. Wallingford. Hoenderken R. (1978) Electrical stunning of slaughter pigs. PhD Thesis. State University of Utrecht, The Netherlands (in Dutch). Huidobro A., Pastor A., LoÂpez-Caballero E. & Tejada M. (2000b) Washing effect on the quality index method

ß 2003 Blackwell Publishing Ltd, Aquaculture Research, 34, 211±220

Aquaculture Research, 2003, 34, 211±220

Is humane slaughter of fish possible for industry? Hans van de Vis et al.

(QIM) developed for raw gilthead seabream (Sparus aurata). European Food Research and Technology 212, 408±412. Huidobro A., Pastor A. & Tejada M. (2000a) Quality index method developed for raw gilthead seabream (Sparus aurata). Journal of Food Science 65, 1202±1205. International patent W.O. 01/95732. (1999) Method and Device for Stunning and Killing Aquatic Animals (ed. by J.W. Van de Vis). World Intellectual Property Organization, 20.12.2001. Kestin S.C., Van de Vis J.W. & Robb D.F.H (2002) Protocol for assessing brain function in fish and the effectiveness of the methods used to stun and kill them. Veterinary Record 150, 302±307. Kestin S.C., Wotton S.B. & Gregory N.G. (1991) Effect of slaughter by removal from water on visual evoked activity in the brain and reflex movement of rainbow trout (Oncorhynchys mykiss). Veterinary Record 128, 443±446. Kuhlmann H. & MuÈnkner W. (1996) Gutachterliche Stellungnahme zum tierschutzgerechten BetaÈuben/ToÈten van Aalen in groÈsseren Mengen. Fischerei und Teichwirtschaft 12, 493±495. Lambooij E., Van de Vis J.W., Kloosterboer R.J. & Pieterse C. (2002a) Welfare aspects of live chilling and freezing of farmed eel (Anguilla anguilla L.): neural and behavioural assessment. Aquaculture 210, 159±169. Lambooij E., Van de Vis J.W., Kloosterboer R.J. & Pieterse C. (2002b) Evaluation of head-only stunning and head to tail electrical stunning of farmed eels (Anguilla anguilla L.) for the development of a humane slaughter method. Aquaculture Research 33, 323±331. Lambooij E., Van de Vis J.W., Kuhlmann H., MuÈnkner W., OehlenschlaÈger J., Kloosterboer R.J. & Pieterse C. (2002c) A feasible method for humane slaughter of eel (Anguilla anguilla L.): electrical stunning in fresh water prior to gutting. Aquaculture Research 33, 643±652. Logtenberg H. (2000) Method for killing of eel. Patent NL C 1008985 (in Dutch). MAFF. (1989) Non sensory assessment of fish quality. Torry Advisory Note 92:6. Torry Research Station, Aberdeen, UK. Mathews G. & Wickelgren W.O. (1978) Trigeminal sensory neurons of the sea lamprey. Journal of Comparative Physiology 123, 329±333. Morzel M. & Van de Vis J.W. (2002) Effect of the slaughter method on the quality of raw and smoked eels (Anguilla anguilla L.). Aquaculture Research (in press). Overmier J.B. & Hollis K.L. (1990) Fish in the think tank: Learning, memory and integrated behaviour. In: Neurobiology of Comparative Cognition (ed. by R.P. Kesner. & D.S. Olson), pp. 205±236. Hillsdale Erlbaum Associates, New York, USA. Robb D.F.H., Wotton S.B., McKinstry J.L., Sùrensen N.K. & Kestin S.C. (2000) Commercial slaughter methods used on Atlantic salmon: determination of the onset brain

ß 2003 Blackwell Publishing Ltd, Aquaculture Research, 34, 211±220

failure by electroencephalography. Veterinary Record 147, 298±303. Robb D.F.H. (2001) The relationship between killing methods and quality. In: Farmed Fish Quality (ed. by S.C. Kestin. & P.D. Warriss), pp. 220±233. Blackwell, Oxford. Robb D.F.H. & Kestin S.C. (2002) Methods used to kill fish: field observations and literature reviewed. Animal Welfare 11, 269±282. Robb D.H.F., O'Callaghan M., Lines J.A. & Kestin S.C. (2002) Electrical stunning of rainbow trout (Oncorhynchus mykiss): factors that affect stun duration. Aquaculture 205, 359±371. Roth B., Veland J.O., Moeller D., Imsland A. & Slinde E. (2002a) The effect of stunning methods on rigor mortis and texture properties of Atlantic salmon (Salmo salar). Journal of Food Science 67, 1462±1466. Roth B., Imsland A. & Moeller D. (2002b) Effect of electric field strength and current duration on stunning and injuries in marked-sized Atlantic salmon held in seawater. North American Journal of Aquaculture (in press). Sigholt T., Erikson U., Rustad T., Johanen S., Nordvedt T. & Seland A. (1997) Handling stress and storage temperature affect meat quality of farm-raised Atlantic salmon (Salmo salar). Journal of Food Science 62, 898±905. Skjervold P.O., Fjñra S.V., éstby P.B & Einen O. (2001). Live-chilling and crowding stress before slaughter of Atlantic salmon (Salmo salar). Aquaculture 192, 265±280. Sùrensen N.K. (1999) `Effect of killing methods for farmed Atlantic salmon (Salmo salar) on quality and storage life, iced in boxes'. Key Note Speaker, Aquaculture Europe, Trondheim, 7±10 August 1999. Spruijt B.M. (1999) Do fish have feelings: a sensitive subject. In: Welfare of Fish (ed. by A.P.J. Raat. & R. van den Bos), pp. 91±98. Tilburg University Press, Tilburg, The Netherland (in Dutch). Tesch F.W. (1999) Der Aal, pp. 18±33. Parey Buchverlag Berlin, Germany. TierSchlV. (1997) Verordnung zum Schutz von Tieren in Zusammenhang met der Schlachtung und ToÈtung (Tierschutz-Schlachtverordnung) vom 3±3 1997. BGBI, Nr 13, 6±3 1997. È nderung der TierssTierSchlV. (1999) Verordnung zur A chlachtung verordnung vom 25±11±1999. BGB1, Nr 54, 10±12±1999. Van de Vis J.W. & OehlenschlaÈger J., Kuhlmann H., MuÈnkner W., Robb D.F.H. & Schelvis-Smit A.A.M. (2001) Commercial and experimental slaughter of eel (Anguilla anguilla L.). effect on quality and welfare. In: Farmed Fish Quality (ed. by S.C. Kestin. & P.D. Warriss), pp. 234±257. Blackwell, Oxford. Verheijen F.J. & Flight W.F.G. (1997) Decapitation and brining: experimental tests show that after these commercial methods for slaughtering of eel, Anguilla anguilla L., death is not instantaneous. Aquaculture Research 28, 361±366.

219

Is humane slaughter of fish possible for industry? Hans van de Vis et al. Wall A.J. (2001) Ethical considerations in the handling and slaughter of farmed fish. In: Farmed Fish Quality (ed. by S.C. Kestin & P.D. Warriss), pp. 108±115. Blackwell, Oxford. Wendelaar Bonga S. (1997) The stress response in fish. Physiological Reviews 77, 592±625.

220

Aquaculture Research, 2003, 34, 211±220

Wiepkema P.R. (1997) The emotional vertebrate. In: Animal Consciousness and Animal Ethics (ed. by M. Dol, S. Kasanmoentalib, S. Lijmbach, E. Rivas. & R. Van den Bosch), pp. 93±102. Van Gorcum & Comp., Assen, The Netherlands.

ß 2003 Blackwell Publishing Ltd, Aquaculture Research, 34, 211±220