There is a general belief that reduced stock numbers on hill areas will be beneficial to the environment. A decrease in stock numbers may lead to farmers opting for lower input systems of hill sheep production in order to reduce costs, to meet the requirements of specific schemes or simply to maintain profitability. A large scale systems experiment has been set up at SAC Kirkton and Auchtertyre Farms to examine the effects of extensification on the ecology, animal production and welfare and the socio-economics of the local community. Within this extensive system, a 4-year trial was undertaken to examine the effects of a limited increase in inputs to twin-bearing ewes which are particularly at risk within such a system.

This paper deals with practical systems where growing pigs or poultry are allowed to self select from two or more feeds. Most commonly a selection between just two feeds has been studied. Only systems are discussed that gave a choice of two feeds where a combination of these feeds gave a combined diet composition that approximately met the nutrient requirements of the pigs or poultry.

Data from singly housed choice-fed animals is most valuable in evaluating their ability to select a suitable nutrient intake. A number of experiments have been published indicating that growing pigs have a poor ability to select different proportions of two feeds. Pigs can discriminate between feeds on the basis of their concentration of crude protein or amino acid. However, the majority of pigs eat predominantly one feed even though a mixture of two feeds would be a more appropriate choice.

There are at least two views about what is the necessary decription of a pig for the purpose of modelling it's performance on controlled feeding. On the first view the potential rate of protein retention is the only descriptor that needs to vary between kinds of pigs, with all pigs assumed to use the same rules to partition scarce resources. On the second view kinds of pigs may differ not only in their potential to grow protein but also in the rules that they use to partition a scarce resource. The purpose of the experiment reported here was to severely test the first view by using two very different kinds of pigs.

The measurement of feeding motivation is of central importance in the study of the factors which control feeding behaviour. Current behavioural models describing the control of feeding behaviour define feeding motivation as a temporary and reversible brain state induced by both internal and external stimuli which result in the tendency to perform feeding behaviour (for a review see Lawrence et al., 1993). One of the most sensitive methodologies to measure feeding motivation is operant conditioning where an individual is trained to perform an operant response (behavioural work) to obtain food reinforcement. An index of feeding motivation can then be obtained by varying the quantity of behavioural work required to obtain each unit of food reinforcement and observing the subsequent operant responding. However, this methodology is flawed because it requires a feed restricted animal to regularly ingest food during the test which results in feedback, affecting the underlying level of feeding motivation. A possible solution to this problem is the use of a second order schedule of food reinforcement which assesses an animal's feeding motivation without allowing it to ingest food. During a second order schedule an individual is required to express an operant response for the repeated presentation of a conditioned stimulus (CS) which has previously been associated with the delivery and ingestion of food, with the schedule terminating with food reinforcement. Using this schedule the quantity of behavioural work required per unit food reinforcement could be gradually increased such that a measure of feeding motivation could be obtained in the period prior to the terminal food reinforcement. Therefore, the aim of this study was to investigate whether the CS presented within a second order schedule had reinforcing properties which were linked to the level of feeding motivation. The experiment formed a part of a larger study which had the ultimate objective of developing a methodology to measure feeding motivation without allowing an individual to ingest food.

Our previous two experiments have investigated the ability of weaned piglets to differentiate successfully between foods on the basis of the protein (CP) content and also on the basis of the lysine content, when offered a choice of two feeds differing in levels of either protein or lysine but with levels of other nutrients kept constant (Dalby et al. 1994a; Dalby et al. 1994b). This experiment was designed to test whether weaned piglets are able to successfully select a diet if both the levels of CP and lysine differ in the foods on offer. This was investigated by offering a choice of three foods containing low and high levels of both CP and lysine, such that it was unlikely that a combination of any two foods would satisfy the piglet in relation to both its CP and lysine requirement.

Recent work (Sinclair et al. 1993) has suggested that synchronising the rate of rumen breakdown and availability of dietary energy and nitrogenous components can increase capture of rumen degradable nitrogen and improve efficiency of microbial protein synthesis. For rumen bacteria to function efficiently they require a supply of both energy and nitrogen sources together i.e a synchronous diet. Nitrogen recycling between the blood and the rumen is a major contributor to the nitrogen economy of the animal and supplies substantial quantities of nitrogen to the rumen bacteria (Kennedy & Milligan 1980). The amount of nitrogen recycled to the rumen is diet dependant. This dependency may not solely be due to the levels of dietary nutrients but also to their rate of breakdown and availability to the rumen bacteria. The predominant mechanism controlling nitrogen recycling between the blood and the rumen appears to be rumen ammonia concentration It has been suggested that the recycling of nitrogen between the blood urea pool and the rumen ammonia pool may be able to compensate for periods of dietary nitrogen deficiency in the rumen due to the asynchrony of availability of the nitrogen and energy components to the rumen bacteria.

The potential benefits of mixing leaves containing high and low levels of condensed tannin (CT) on N utilisation in sheep was studied at the Centro Internacional de Agricultura Tropical (CIAT), Colombia.

Supplements of Cratylia argentea (Ca) (tannin free), replaced with 0, 25, 50 or 100% Flemingia macrophylla (Fm) (25.1 g CT/kg DM) were fed as 40% of total ration DM (85g/M0·75), the remainder coming from Brachiaria dictyoneura (Bd). Treatments were offered in a replicated 4x4 latin square to eight male African Hair sheep (mean liveweight 23 (s.e. 1.5) kg), fistulated at the rumen and proximal duodenum (T-piece). Total collections of urine and faeces were used to estimate apparent N retention. Ammonia concentration was measured in rumen liquor samples taken hourly over 24 hours, while duodenal flows were estimated from spot samples using Ytterbium chloride as a single phase marker.

In an early lamb production system feed costs and age at slaughter are big factors in determining profit. Farmers are being forced to become more efficient producers as costs of production continue to rise and farm product prices fail to keep pace with inflation. The aim of this experiment was to evaluate the performance of lambs fed concentrates ad-libitum to those fed to a maximum restricted level of 0·5 kg/hd/day.

The term biotechnology has been defined as the application of biological organisms, systems or processes to manufacture and service industries (Anon, 1980) and is, therefore, more than the application of ‘genetic engineering’ techniques alone. The potential application of biotechnology to the agricultural livestock industry includes many wide ranging areas: animal health; breeding; livestock production; livestock nutrition and the nutritive value of feeds. The role of biotechnology in animal nutrition and feeding is of particular importance since feed costs account for a significant proportion of the total variable costs in any livestock production system (e.g. milk, meat). The potential implications of biotechnology in animal nutrition has been reviewed by Armstrong (1986) and includes references to the improvement of the nutritive value of feeds through, for example, genetic manipulation of feed sources (cereals), appropriate supplementation and the use of biological inoculants with or without enzymes as silage additives and to improvements in the ability of the animal to obtain nutrients from feeds through the addition of enzymes to feeds and modification of rumen microbes through genetic engineering.

Results from a series of recent experiments involving superovulated ewes demonstrate the important influence of nutritionally-induced alterations in preovulatory progesterone concentrations on the subsequent in vivo and in vitro development of their fertilized ova (McEvoy et al, 1993 and 1995; Creed et al, 1994). In essence, these show that high-plane feeding can suppress preovulatory progesterone concentrations to such an extent that the subsequent development of the ova is impaired both in vivo and during in vitro culture. An important practical question however remains unanswered in that no attempt has been made to study the effects of dietary energy concentrations, as opposed to plane-of-nutrition, on progesterone concentrations and ovum development. As a result, recommendations regarding which energy sources should be used as supplements to pasture around mating time are a matter of conjecture. Furthermore, in arid environments, roughage feeds are often in short supply and therefore command a much higher price per unit of energy than concentrate diets. Under these conditions it is not unusual to feed all-concentrate diets at mating, yet there are no published data for their effects on ovum development and embryo survival.

Earlier work has shown that newly-housed, previously-pastured lambs are reluctant to eat straw. This experiment monitored intake and diet-induced behaviour patterns in newly-housed lambs with the aim of determining the effect of ammoniation on rate of adaptation to straw diets.

Sixteen, 35 kg lambs were selected from a flock of seven month old, Suffolk cross Mule lambs which had been pastured from weaning and had no previous experience of straw. Lambs were housed in October in individual, continuously- lit, open-side pens and fed untreated or ammonia-treated (100 1/t of 320 g/kg ammonia) winter wheat straw. Ammonia treatment was by the stack method (Sundstol and Coxworth, 1984) and the straw offered was collected in 20 kg wads every other day from stacks kept in an open-sided barn. A completely randomised design with eight replicate lambs per treatment was used to compare straw types. Sufficient straw was offered to allow daily refusals of at least 300 g/kg offered during the 21-day trial. A barley-based supplement was fed at 240 g dry matter (DM)/d. Nitrogen contents of both straw types were monitored with a LECO FP-228 automatic nitrogen determinator while the total content of neutral and acid detergent fibres were analysed using the method of Goering and Van Soest (1970). Thein vitro dry matter digestibility (DMD, Tilley and Terry, 1963) of the straw offered was also monitored. A concurrent behaviour study involved recording the activity (eating, drinking, idling or ruminating) every 5 minutes for 9 h on day 1 and 12 h on days 8 and 16 of the 21-day trial.

Predictive models for estimating the nutrient requirements of ruminant livestock have become increasingly complex in their approach, and implicitly they have a need for accurate characterisation of feed degradability kinetics (AFRC, 1992; Russell et al.,1992; Sniffen et al., 1992; Fox et al., 1992). The rumen in situ technique has been adopted as the standard method (AFRC, 1992) for the characterisation of nitrogen degradability in feeds for ruminants. Recent ring tests (Madsen and Hvelplund, 1994) and reviews (Huntington and Givens, in press) have highlighted the need for improved standardisation of the procedure. The aim of this experiment was to investigate ruminant species differences and to determine the effect of anchor weight and incubation sequence on dry matter (DM) disappearance of hay, soyabean meal and fishmeal in situ.

The results of a previous experiment showed that when the amount of compound feed fed to crossbred ewes during late pregnancy and early lactation was reduced, being replaced by feed-blocks, animal performance was maintained and that the cost of supplementary feeding was similar. In the same experiment, when compound was replaced totally by feed-blocks plus Scotmol® (a blend of molasses and pot ale syrup), feed costs were considerably reduced, but animal performance was not acceptable (Merrell and Marsh, 1994). In the current experiment done at ADAS Redesdale in 1994 the proportion of compound feed replaced by feed-blocks was increased compared with the previous experiment, in an attempt to reduce feed costs, and the feasibility of replacing part or all of the copmound with cane molasses and fishmeal was tested.

While the benefits of using maize silage and ensiled super pressed pulp in the diets of lactating cows has been well researched, little work has been done on the inclusion of these feeds in the diets of pregnant ewes. The objective of this experiment was to compare the effects of the nutritive value of grass silage, maize silage and a maize silage/ensiled super pressed pulp (60:40) mixture, when offered to mature twin bearing ewes in late pregnancy on ewe and lamb performance.

There is considerable scope for modifying the extent of fermentation during silage production by selecting an appropriate silage additive, with chemicals such as formic acid restricting fermentation and/or fermentable carbohydrates stimulating the natural fermentation in the silo. Both approaches can achieve satisfactory preservation but can produce silages of quite different chemical compositions. The objective of this experiment was to compare the pattern of rumen fermentation and digestibility of a silage, which had 50 kg of molassed sugar beet pulp (MSBP) applied at harvesting, with those of an untreated silage and a silage treated with formic acid.

The pattern in which ruminants consume their daily allocation of forage may affect their total daily intake and the pattern of supply of energy and protein to the micro-organisms in the rumen. Synchronizing energy and nitrogen supply to the rumen has been demonstrated to improve the efficiency of microbial protein synthesis (Sinclair et al. 1993, 1995). However, it remains unclear whether animals, when offered a feed with a predicted asynchronous release of nutrients, will alter their pattern of intake to achieve a synchronous supply. In addition, the provision of a concentrate may alter daily intake, hourly intake and the pattern of release of nutrients in the rumen. The objectives of the current experiment were to examine the effects of level of supplement on the pattern of intake and the predicted release of nutrients in the rumen in sheep fed grass silage.

Methane production by ruminant animals is an important source of methane to the atmosphere and is considered an important source to more accurately estimate and to control. It has been known for some time that, in ruminants, diet type and composition can influence methane production, yet there is little quantitative information for contemporary diets used in the UK. Attemps to predict methane production from either diet chemical composition or digestibility have not been successful (Moss and Givens. 1990; Johnson et al., 1991). In order to estimate more accurately the amount of methane produced by a ruminant on a given diet, it is essential to establish whether there are interactions of various substrates with forage and to relate this to the rumen environment. Previous work showing interactions has used mainly hay-based diets (Blaxter and Wainman. 1964): as grass silage is now the major forage in the UK, work with this forage was undertaken.

Increasing concern over the role of greenhouse gases in global warming has led to a renewed interest in models for predicting methane production in ruminants. For economic reasons it would be preferable if such models could be derived from data obtained in smaller ruminants. The objective of this project is to establish whether differences in methane production exist between sheep and cattle and to measure the magnitude of these differences under various dietary situations.

Protein supplementation has a positive effect on milk production, particularly in ewes on negative energy balance (Gonzalez et al., 1985). The response has been associated either to an increase in the movilization of body reserves (Whitelaw et al., 1986, Hadjipanayotou et al., 1988) or to an enhanced efficiency of energy utilization for milk synthesis. On the other hand, the response has been shown to be higher with fish meal than with other protein sources (Oldham, 1984). This work aimed to study the mecanisms involved in the response to fish meal supplementation, in lactating ewes on negative energy balance.

The sward height could determine the intake of the herbivores and, at the same time, to affect to sward composition (Morris et al., 1994). If the foraging strategies is different among species could be also different among breeds of sheep (Milne, 1991).

The objective of this work is to know the effect of sward height and sheep breed (churra and merina) on sward composition under irrigated and continous grazing system.

The experiment was carried out according a factorial design 2x2x5, defined by:

2 sward height (SH): low (LSH:<3.5 cm) and high (HSH:>6.5 cm)

2 Spanish breed of sheep: Churra and Merina

5 sampling periods: 1 (29.07), 2 (20.07), 3 (24.08), 4 (21.09) and 5 (19.10)

The SH was recorded three times a week and they were adjusted using variable stocking rates with spares similar ewes. Available dry matter (kg/ha) was 318 and 943 for the low (3.7 cm) and high (6.5 cm) sward height, respectively.