A contingent valuation survey of 2000 people in Great Britain was conducted to find out about people's concerns about farm animal welfare and their willingness to pay to support legislation to phase out the use of cages in egg production in the European Union (EU) by 2005. The survey found that respondents were concerned about a range of farm animal welfare issues, and that the majority supported the proposed legislation and stated that they were willing to pay a significant increase on the current price of eggs to support it. People's willingness to pay to support the legislation would appear to outweigh the extra resource costs of producing eggs without using cages. However, possible sources of bias in the survey and the validity of the willingness to pay measure estimated are discussed. It is concluded that such measures must be interpreted with caution, but can be useful indicators of people's concerns about animal welfare issues and their preferences for public policies on animal welfare.

In nature, mice live in burrows with nest chambers where they breed and may hide from predators. In the laboratory, a shelter or refuge is an easily applicable form of enrichment which may enhance the welfare of laboratory mice by giving them more control over their environment. Six nest boxes made of different materials were evaluated in a preference test with male and female mice of two strains (C57BL/6J and BALB/c). In general, mice showed a preference for cages with a nest box made of grid metal as compared to clear or white perspex nest boxes, or no nest box. They also showed a preference for a cage with a nest box of perforated metal as compared to nest boxes made of grey PVC or sheet metal, or no nest box. When offered a nest box with one open side or a nest box with two open sides, most mice preferred the nest box with one open side and were observed to lie in it with their heads directed towards the opening.

The results of this study show that nest boxes may be used for enrichment purposes in caged mice, although it is not yet entirely clear what are the main features influencing the animals’ preferences. When providing nest boxes as shelters, the structure and design of this type of enrichment should be taken into account, because these may have an effect on the social structure of groups of mice.

The preferences of growing pigs for substrates were investigated by giving small groups of pigs a choice between two substrates in each test. The seven substrates examined were concrete, mushroom compost (spent), peat, sand, sawdust, straw and woodbark. Thirteen comparisons of pairs of substrates were tested with four replicates of each comparison. Eleven-week-old pigs (in groups of six) were placed in specially designed choice pens where they had access to two different substrates. The pigs were allowed to habituate to the pen for 1 week and at the end of week 2 the substrates were swapped. In weeks 2 and 3 the time spent by the pigs in each substrate was recorded. Peat, mushroom compost and sawdust were preferred most, with sand next and woodbark and straw being preferred only to concrete. It is suggested that growing pigs may be attracted to substrates which are similar in texture to earth.

Investigations of free-living wild animals often involve capture, restraint or other forms of manipulation. There is a need, on both scientific and welfare grounds, to develop and to use less invasive techniques, especially for the assessment of health. Already some such procedures exist, ranging from observation of clinical signs to the laboratory examination and analysis of faeces and other naturally voided samples.

Minimally invasive and non-invasive health monitoring of vertebrate animals is outlined and examples are given of samples from mammals, birds, reptiles, amphibians and fish that can provide useful information.

To assess the welfare of red deer (Cervus elaphus) confined at pasture or in indoor housing over winter, behaviour, productivity, skin damage and adrenal response to ACTH challenge were measured in six groups of eight weaner hinds over 91 days from June to September 1990 in Otago, New Zealand. The hinds were confined either indoors (I), indoors with daily exercise (IE), or outdoors (O); (n - 2 groups to each treatment). All groups were fed concentrate ad libitum plus 100g lucerne head−1 day−1.

Indoor confinement was associated with a greater incidence of nosing/chewing other hinds, aggression, chewing of the enclosure, and closer distances between individuals, compared with outdoor confinement (P < 0.05). Ad libitum provision of hay over a 2-week period reduced the incidence of chewing of indoor enclosures (P <0.01). Weight gain was greater for indoor groups than outdoor groups in August and September (P <0.05) and overall weight gains for indoor groups (from two weeks into the study, until the end) were higher for the exercise treatment (P <0.05). Intake of concentrates did not differ significantly between treatments. Skin damage was greater for indoor than outdoor groups (P <0.05), and positively related to weight gain (P <0.01) and receiving aggression (P <0.01), which in turn was negatively related to liveweight (P < 0.001). A negative relationship was found between pre-challenge levels of plasma Cortisol and the number of aggressive interactions received (P < 0.05). Pre-challenge Cortisol was greater for IE than I (P < 0.05), and the increase in Cortisol post-challenge was greater for outdoor groups than indoor groups (P <0.01). Conclusions were that indoor confinement had a positive effect on weight gain, but increased aggression and skin damage, indicating that the deer were compromised socially. Provision of ample forage reduced chewing of the walls. The slightly greater weight gain in IE compared with I deserves further investigation.

The behaviour of a captive pair of clouded leopards was studied during a series of manipulations in order to devise a safe method of introduction for mating purposes. Manipulations consisted of allowing each individual unrestricted access to the other's outdoor enclosure, initially in the absence of the other individual, but culminating in joint access. Dominant activities involved sitting, lying, grooming, and adopting a low profile amongst the vegetation. The female tended to be more arboreal than the male, although both cats spent most of their time on the ground. The male marked various sites by foot-scrubbing (1-4% occurrence), which involved shuffling urine into the ground using his hind feet. This was less common in the female (<1%). Male foot-scrubbing was most frequent on introduction nights, and in the female's enclosure. Both sexes exhibited cheek-marking behaviour, although it was more common in the male. The frequency of male cheek-marking increased in response to urine production by the female. Observations during introductions suggested that the male may assess the female's reproductive condition by stimulating her to urinate. The most marked changes in the behaviour occurred between control and introduction nights. The larger male took the initiative, and the female appeared extremely wary of his presence, striking out with her claws if he approached too closely. Although the individuals did not mate during the introductions, the method of gradual acquaintance through an experimentally induced overlap of ‘home ranges’ was effective, as the female was not injured even though the male had a history of aggression.

This study compares cortisol concentrations in plasma and milk over a time-span of 1-4h following injection of an adrenocorticotropic hormone (ACTH) or a physiological stressor. Its aim was to characterize the usefulness of milk cortisol concentrations as short-term measures of acute stress.

Experiment 1: three groups of lactating cows (Groups B - D; n = 5, each) were injected with ACTH1-24 at 4, 2 and 1h before milking, respectively, so that each experienced a similar period of elevated adrenocortical activity, but presented at milking with varying plasma cortisol concentrations. Another group (Group A; n = 5) was a saline-treated control. Mean plasma cortisol concentrations at milking were 7, 8, 24 and 56ng ml−1 for Groups A, B, C and D, respectively. Mean cortisol concentrations in milk were similar for Groups A and B (1.2 and 0.5 ng ml−1, respectively), higher in Group C (2.4ng ml−1), and greatest in Group D (11.7 ng ml−1; p <0.001).

Experiment 2: lactating dairy cows (n = 15) were injected with ACTH, transported by truck, or blood-sampled only (control) during the 2h before milking. Mean plasma cortisol concentrations at milking were 6, 20 and 72 ng ml−1 following control, ACTH and transport treatments, respectively; mean concentrations of cortisol in milk displayed a similar pattern (1.1, 2.4 and 12.0 ng ml−1, respectively; p <0.001).

Milk cortisol concentrations were highly correlated with plasma levels at milking, but did not reflect those situations where, following a period of elevation within the previous 4h, plasma cortisol concentrations had returned to basal levels. Concentrations of cortisol in foremilk and composite milk were highly correlated, but the mechanisms of cortisol flux may differ between these two compartments. Milk cortisol concentration can be a useful indicator of responses of lactating cows to acute stressors which act up to 2h before collection of samples.

Automatic broiler catching machines have been developed for several reasons. Manual catching is expensive, often rough and may cause injury to the birds. Apart from animal welfare considerations, the demand for good quality meat means that rejection rates must be kept at a low level. The poor working conditions for manual catching teams are also an important factor. Only a few scientific studies on the effects of catching machines on bird health and welfare have been published. In this study, the carcase rejection rates in relation to manual and mechanical catching were compared at a poultry abattoir following a change of loading and unloading systems. The effect on the level of birds found ‘dead on arrival’ at the slaughterhouse initially varied considerably, regardless of catching method. During the last 3-month period of the study however, mechanically caught flocks showed significantly higher frequencies of dead birds on arrival than manually caught flocks. During the same last period, when running-in problems should be regarded as solved, the prevalence of bruises was also still significantly higher in the mechanically caught flocks, whereas the prevalence of fractures did not differ significantly between the two groups. This suggests that there are still opportunities for further improvements of the machine, although in Sweden it has now been accepted for commercial use in accordance with Swedish animal welfare legislation, under the supervision of the National Board of Agriculture.

As a contribution to the ongoing work aiming to improve welfare in farmed silver foxes (Vulpes vulpes), the present study attempted to investigate stress-induced hyperthermia (SIH) as a physiological indicator of fear or anxiety in this species. Measuring rectal temperature (Trec) served as a stressor and immediately elicited SIH. Then, the foxes received either the anxiolytic drugs, diazepam or buspirone, or sterile saline. Trec in the foxes treated with diazepam and buspirone was significantly lower 90min after treatment. The results indicate that SIH induced by handling was attenuated by anxiolytic drugs, which supports the hypothesis that anxiety contributes to the development of SIH.