The causes and timing of piglet mortality were studied in different farrowing systems. In the first experiment 198 litters were recorded in three systems, two of which allowed the sows to move freely, and the third restricted them in conventional crates. More piglets were weaned from the conventional crates than from the open systems and they grew more quickly. More than half the liveborn mortality occurred during the first four days after parturition. In the open systems, 17 per cent and 14 per cent of the piglets born alive were crushed, compared with only 8 per cent in the crates. In the second experiment, 29 sows and litters were studied in detail in a communal pen system during the first seven days of lactation. Three-quarters of the liveborn mortality was due to crushing. The total number of piglets dying per litter, including stillbirths, was significantly associated with the total litter size and the sow's parity. The percentage liveborn mortality was significantly associated with the parity and body length of the sows and with the within-litter variation in the birth weight of the piglets. Individual birth weight was closely associated with percentage survival. Only 28 per cent of piglets weighing less than 1.1 kg at birth survived to seven days.

No abstract

Previous research showed that pigs reared in substrate-impoverished conditions performed a smaller proportion of their total behavioural repertoire in their home pens (showed lower behavioural diversity), than pigs reared in substrate-enriched conditions. This study examined whether these differences were the result of fundamental changes in behavioural organisation. A T-maze task was used to test the hypothesis that substrate-impoverished pigs are prone to develop fixed, unvarying behaviour which may underlie their reduced behavioural diversity. They were predicted to be poorer at reversing previous response patterns in the maze, and less able to alter their behaviour in response to a novel (distracting) stimulus. Female pigs were housed singly for five months in substrate-impoverished pens with bare concrete floors (N = 10) or substrate-enriched pens with straw and other foraging material (N = 10). The pigs were then trained to negotiate a T-maze to reach a food source. There were no differences in responses to a distracting stimulus in the start arm of the maze but, contrary to expectation, substrateenriched pigs were less able to change their behaviour when the route to food was switched. Thus, the hypothesis was not supported. During training trials, substrate-enriched pigs moved fairly rapidly to the food while substrate-impoverished pigs spent more time investigating the maze; their motivation to 'explore' the maze appeared to override their interest in food. The rapid, food-directed behaviour of the substrate-enriched pigs probably became more fixed and routine-like than the more exploratory behaviour of the substrate-impoverished pigs. The apparent importance of exploration to pigs reared in substrateimpoverished environments suggests that such conditions provide inadequate stimulation.

Although the physiological and behavioural changes that can indicate poor welfare are generally agreed upon, using these measures in practice sometimes yields results that are hard to interpret. For example, different types of measure may suggest quite different things about an animal's welfare. Such contradictions are often due to the differing properties of the variables being measured. How each variable responds to a stressor can be affected by several factors - the type of unpleasant stimulus to which the animal is exposed; when and for how long exposure occurs; the animal's psychological state, eg does it feel that it is in control?; and the time at which the measurement is made, relative to the stressor. Typical responses also often differ between species and between individuals, and may even change in a single individual over time. Furthermore, some responses used to assess welfare lack specificity: they can be elicited by neutral or even pleasant events as well as by aversive ones. Appreciating these factors is vital when designing experiments, when choosing what to measure along with each welfare variable, and when interpreting results. Even after taking these factors into consideration, interpreting a result can still be difficult. One approach then is to consider the effects on welfare of the changes measured, eg if there is immunosuppression, does the animal succumb to disease? Another is to use the animal's behaviour to indicate its preference for, or aversion to, particular environments. Ultimately, however, interpreting welfare measures involves subjective judgements which will be influenced by the nature of our concern for the animal under consideration. By raising these problems, we hope that this review will highlight and clarify the apparent contradictions that sometimes emerge in scientific studies of animal welfare, and help researchers improve the designs of their experiments for the benefit of the animals concerned.

No abstract