Backfat levels in pigs have declined rapidly in the last decade and are approaching an economic optimum. As optimal levels of backfat are achieved this has two important implications for selection objectives. First, the relative emphasis of selection will shift to other traits, particularly lean growth rate. Second, there will also be a move towards ad libitum feeding on performance test to allow expression of appetite and to increase response to selection for lean growth rate. The effect of very lean pigs coupled with ad libitum feeding may result in changes in the genetic and phenotypic relationship between growth rate and fat depths. Reported genetic and phenotypic correlations between growth rate and backfat range from +0.3 on ad libitum feeding to -0.3 on restricted feeding. In most performance testing regimes the key relationship is between growth rate and ultrasonic fat depths which in contrast to other economic traits are often unfavourably correlated.

Genetic improvements and developments in production methods have made dramatic changes in the composition of British pig carcasses over the last ten years. As a result, equations for predicting lean concentration are likely to need frequent revision. The first Stotfold Trial was designed to obtain a comprehensive modern data set with which production efficiency, carcass composition and meat quality could be examined for different breed types, sexes and feeding regimens. It therefore provides a valuable database with which to evaluate predictors of carcass lean. The main objective of this paper is to examine the stability of the regression equations predicting carcass lean content from carcass weight, P2 and lean content of sample joints among different populations, sexes and feeding regimens. Two different aspects of prediction were studied. The first was that of carcass grading, the estimation of lean concentration from P2 (measured with an Optical Probe) and carcass weight. The second was that of the estimation of total lean concentration from the dissection of sample joints and from the removal of total subcutaneous fat in the side. The accuracy of these estimators was also examined.

For the pig breeder there are many important selection objectives - optimum progress can be made if sires and dams are selected for different objectives. In dam lines some emphasis on maternal traits is practicable while maintaining slightly higher fat levels than in sire lines. Meanwhile lean growth and low fatness can be emphasised in the terminal sire lines. This gives the flexibility to fulfil different end markets and weights. Selection accuracy in such programmes is often improved by the addition of detailed carcass information from slaughtered sibs or other relatives of selection candidates.

For the commercial producer, classification schemes differentiate payments on the basis of backfat and in some cases, lean measurements. However, despite the fact that future payments are likely to place greater emphasis on leanness, there is no differentiaton for breed or genotype.

The purpose of this investigation was to follow up a previous study (Walters and Simpson, 1988) to determine whether lean prediction in the carcass depends on slaughter weight and genotype, to identify the best predictors of leanness and to estimate multiple correlation coefficients for predicting lean in carcass from lean in sample joints.

A revised EC Pig Carcase Grading Scheme was introduced on 1 January 1989. It applies at all abattoirs slaughtering over 200 pigs a week. Carcasses are graded on the basis of -their estimated lean content. The estimate is made from prediction equations approved for use in each country and based on objective measurements, normally including subcutaneous fat depth, muscle thickness and carcase weight. Each carcass is placed in one of five (or optionally six) grades, the grades representing steps of 50g/kg in estimated lean content.

Three British breeds, Large White, Landrace and Welsh are used widely in producing crossbred sows for commercial slaughter pig production in Zimbabwe. In most cases crossbred sows are mated either to Duroc or Hampshire to produce slaughter progeny. This study was undertaken to evaluate the performance of these five breeds under test station conditions.

The records for this study were obtained from the central test station. The data included 1513 centrally tested pigs over the five year period 1984-88. The traits analysed were daily gain (DG) g, food conversion ratio (FCR) between 35-86 kg, k backfat depth at 86 kg (BF) mm, conformation-score (CS) and leg-score (LS). The data for CS and LS were on a 50-point scale, taken at the end of the test. Animals were fed on a restricted scale during performance testing. The data were analysed by least-squares procedures. For all the traits fixed effects considered in the analysis were season, year, sex, breed, age at start of the test and interactions as appropriate. The season according to month at start of performance test was classified as cold (April-June), dry (July-September), early Summer (October-December) or late Summer (January-March).

Feeding strategies for the lactating sow must aim to achieve high piglet survival and growth rates and maximise subsequent reproductive performance of the sow. Feeding less frequently and feeding ad libitum may be ways of reducing disturbance and crushing of piglets, enhancing milk yield and maintaining sow body condition to enhance subsequent performance. High protein diets may encourage sows to utilise body fat reserves for milk production without compromising rebreeding. This experiment was designed to investigate these possibilities in commercial practice.

100 Camborough hybrid sows were allocated to one of 4 dietary treatments:

• 1)ration fed with the daily allocation divided between two feeds

• 2)ration fed with the total daily allocation given in one feed

• 3)fed ad libitum on the standard diet (CP 150 g/kg)

• 4)fed ad libitum on a high protein diet (CP 180 g/kg).

The feeding scale for the ration fed sows, based on commercial farm practice, was as follows: 2.5 kg on days 1-3, 3.5 kg on days 4-5, 5.0 kg on days 6-9 and 7.0 kg/day thereafter.

The experimental diets were based on barley and soya, with the higher protein diet obtained by substituting an additional 75 g/kg of barley with soyabean meal.

Both level of feeding and dietary composition have been demonstrated to influence bacterial colonisation and diarrhoea in the weaned pig. Specifically, Bertschinger et al (1979) demonstrated that diets low in protein and energy and high in fibre content could protect the weaned pig against challenge with live pathogenic E. coli. However, pig growth was so compromised that it proved to be unsatisfactory for commercial practice. Miller et al (1984) demonstrated that a diet in which the protein (casein) had been hydrolysed to constituent amino acids or peptides, thus markedly reducing its antigenicity, also could prevent post-weaning diarrhoea and minimise associated gut changes. They therefore, suggeted that one of the factors which may contribute to the aetiology of post-weaning diarrhoea is a transient hypersensitivity to dietary antigens prior to tolerance development.

The aim of this present work was to attempt to combine these two approaches yet achieve acceptable piglet performance, the long-term aim being to design weaner diets which effectively control post-weaning diarrhoea on commercial farms without resulting in the use of antibiotics either prophylactically or therapeutically.

Hyperinsulinaemia, which is associated with obesity and type II diabetes (Gama and Marks, 1989), may be due in part to an overactive enteroinsular axis. The gut hormone GIP is a potent stimulator of insulin secretion and is currently thought to be one of the main components of the enteroinsular axis. Exaggerated GIP responses have been observed after a mixed meal in obese subjects (Hampton, Kwasowski, Tan, Morgan and Marks, 1983), and a high-fat diet increases fat-stimulated GIP secretion in rats (Ebert, Willms, Brown and Creutzfeldt, 1976) and glucose stimulated GIP secretion in man (Morgan, Tredger, Hampton, Kwasowski, Wright, Dunn and Marks, 1983). This experiment was designed to investigate the effects of high-fat diets on the development of the biochemical parameters which are characteristic of obesity.

The daily rate of lean deposition achieved by growing pigs is a function of a wide range of factors including genotype, gender, liveweight and intake of energy/protein. The review of ARC (1981) highlighted the controversy surrounding the effects of liveweight and energy intake on lean deposition. Recent publications suggest that there are interactions between these factors and also with genotype. In contrast to the linear/plateau relationship between energy intake and protein deposition proposed by Whittemore and Fawcett (1976), Campbell and Taverner (1988) observed a linear response in protein deposition up to the highest energy intake achieved, with pigs of improved genotype. The slope of the relationship was much greater than that observed in previous studies (ARC 1981). In a recent experiment, McCracken and Rao (1989) have shown that high-lean pedigree boars can achieve protein deposition rates as high as 200 g/d over the liveweight range of 33 to 88 kg. At present there is no published information on the response of such pigs to energy intake though the low rates of fat deposition observed suggest that energy intake could be limiting protein deposition. The experiment described below was designed to measure the response of protein deposition to energy intake at a series of liveweights between 33 and 88 kg. The diet and the treatments were chosen to obtain a wide range of energy intakes above and below those observed with dry, pelleted diets.

The fatty acid composition of meat has important effects on its succulence and flavour. The neutral lipids of muscle, are mainly energy stores resembling adipose tissue and affect the succulence of meat. However, the phospholipids, which are structural components of muscle, are important in the development of the species specific cooked meat flavour and odour. We have investigated the effects of feeding diets differing mainly in their linoleic acid content on the fatty acid composition of the lipids of M. Longissimus dorsi (LD) of pigs.

Cholesterol plays an important role in the transport of lipid throughout the body. Moreover, reports have linked the turnover of cholesterol with obesity (eg Nestel, Schreibman and Ahrens, 1973). Recent studies have demonstrated wide variations in serum cholesterol in response to hypercholesterogenic diets in growing pigs (Shutler and Low, 1988). These variations in response appear to have a genetic basis and may be due to variations in turnover rates. The aim of the present study was to determine whether the measurement of serum cholesterol in growing pigs, either in their normal state or after being given a dietary cholesterol challenge, could be used in the prediction of performance or carcass quality.

The role of stockmanship in relation to the wellbeing of farm animals has been emphasised in the UK Codes of Recommendations for the Welfare of Farm Livestock (MAFF, 1983). Moreover, previous research work has demonstrated important effects of good and bad stockmanship not only on welfare but also on growth, lactational and reproductive performance of pigs and other farm livestock (Hemsworth et al, 1987). There is a need, therefore, to establish the factors which motivate and demotivate stockpeople since the degree of job satisfaction is likely to have a considerable influence on the attitude and performance of stockpeople and on their empathy with the animals in their care. With this objective in mind, a questionnaire was designed to investigate the aspects which could have an influence on the job satisfaction of the stockpeople involved in pig production.

Pig producers must continuously strive for improved productivity and efficiency in order to remain competitive. The effect of sow culling rate on overall herd productivity is well documented in the scientific literature (Kroes and Van Male 1979, Dagom and Aumaitre 1979, Dujkhuizen 1986). Recent indications, however, are that there is still much room for improvement in the efficiency of sow culling policy. The objectives of this work were a) to determine the incidence of and reasons for culling in a modern 800 sow commercial pig farm using Large White x Landrace sows, and b) to attempt to develop a more objective basis for culling on poor litter performance (essentially low numbers born).

The objective of the present study was to investigate the development of hydrolase enzyme activity at six sites along the small intestine of the pig before and after weaning and the influence inclusion of a probiotic preparation had upon this. The enzymes studied lactase, dipeptidase and tripeptidase are primarily associated with the brush border of the enterocyte cell and expression of hydrolase activity has been shown to be affected by manipulation of the microbial population of the gut.

The computerised spreadsheet model which forms the basis of this study, describes two representative Welsh Dairy farms. Its purpose is to critically evaluate the economic effects of technological and management changes within the Welsh dairy sector, and to assess their applicability at a farm level.

To provide the physical and financial parameters for the model, a statistical analysis of the physical and financial data on Welsh dairy farms held on the Welsh Farm Management Survey Database was performed. From this information two representative farms were modelled. Farm Type 1 (S) is a 32ha holding with a milk quota of 129,000 litres, and Farm Type 2 (L) a 67ha holding with a quota of 407,840 litres. Both rear their own heifers as well as some store cattle, and have small flocks of breeding ewes. Farm type 2. also grows feed barley for use on the farm.

Animal behaviour is increasingly seen as a useful tool in animal production. However, behavioural observations are time costly, and it is difficult to ensure precision and repeatability of measurement.

An alternative to observation is to use electronics to automatically record behaviour and two such applications are illustrated:

The first system developed at Edinburgh records drinker and chain-related activity, two common abnormal behaviours in pregnant sows. Water flow meters record drinker use. Chain behaviour is monitored by supplying each sow with a vertically hanging chain to which is attached piezo-electric ‘wire’. Electrical activity is generated from the flow meter or the piezo-electric wire as the sow manipulates the drinker or the chain, and is read by a micro-computer at 60 seconds intervals. The system estimates water use and circadian patterns of drinker and chain activity for 16 sows. Applications include recording the development of these two behaviours in response to variations in variables such as feeding regime.

With the widespread use of AI on many commercial dairy farms, the efficient detection of oestrus has become a subject of increasing importance. Of all the physiological variables that have been investigated as predictors of oestrus, body temperature monitoring either on its own or in conjunction with other physiological variables, has shown promise mainly because of the ease with which it can be adapted as a practical on-farm oestrus detection system. However, oestrus detection rates by body temperature monitoring rarely exceed 70-80%, and often 10-20% of non-oestrus cows are wrongly diagnosed as being in heat. If the percentage of false positives is to be minimised, then identifying those factors that affect body temperature is important. Equally important, is an understanding of the way and the extent to which each factor affects body temperature. The more complete these two pieces of information, the more confident can one be of diagnosing an oestrus cow correctly. The aim of the present study was to investigate whether the behaviour of a dairy cow affects her body temperature, and what light this information can throw on the use of body temperature as a predictor of oestrus.

There is a need for objective measurement of the carcass composition of sheep to apply to live animals (selection) and to the carcass itself (classification). Ultrasonic scanning of live sheep has generally produced disappointing results, and tissue depths measured by automatic probes on carcasses have not been as precise as the best visual scores for fatness. The application of the velocity of sound (VOS) technique to lamb carcasses in a previous study (Fisher and Page, 1986) measured composition at a hindlimb and neck site but was not as precise as fat scores in predicting lean proportion. A recent patent describes a method for determining the speed of ultrasound through soft tissues in the vicinity of bone (Miles, Fursey and Page, 1989), allowing measurements to be made across the back through M longissimus thoracis et lumborum and between the spinous processes of the vertebrae. This study uses that method to examine VOS measurements made at sites along the vertebral column in live sheep and carcasses.

Unique and secure identification of livestock from birth through to the carcass in the abattoir will become a key feature in many areas of livestock production: disease eradication; quality assurance schemes; pedigree breeding; and producer feedback from abattoirs. There have been many attempts at adapting traditional methods of animal identification to provide a measure of automatic recognition. None has been as successful as the application of radio frequency transponders which transmit a unique code when triggered by a suitable reader.

The meat industry needs a skilled workforce to meet the challenges of the 1990's. The need is particularly acute in computing and information technology (IT) because people with the necessary skills are becoming notoriously difficult to find, even for the most technologically advanced and high-salaried industries. The expected fall by 20% in the number of young people in the national labour force between now and 1995 will exacerbate these difficulties, as will the poor perception that potential employees have of the meat industry.