The effects of spring top-dressed applications of broiler litter, pig slurry and cattle slurry on winter wheat and winter barley yield and quality were investigated on a field scale at three UK sites between 1992 and 1994, using commercially available application equipment. Broiler litter was applied at rates ranging from 5·3 to 8·8 t/ha, and slurries from 54 to 89 m3/ha. Few practical problems were encountered when spreading broiler litter, but when spreading cattle slurry there was some crop damage and soil compaction from the tanker wheelings adjacent to tramlines. Coefficients of variation for manure spreading using commercial spreaders ranged from 20 to 32%. Spring applied manures increased yields of winter wheat and winter barley, and lowered optimum inorganic fertilizer nitrogen (N) rates. When the N efficiency of the manures was compared to that of inorganic fertilizer N, broiler litter N efficiencies ranged from 10 to 49%, cattle slurry was c. 30% and pig slurry c. 50%. The experiments demonstrated that poultry litters and slurries can be applied successfully to growing cereal crops in spring as part of an integrated policy for N supply.

The long-term effect of fertilizers and an integrated nutrient supply system was studied in a rice–rice sequence for 8 years at Hyderabad in India from 1988/89 to 1995/96. Nutrient uptake and seed yield improved with increase in fertilizers up to the recommended dose of 120, 60, 40 kg/ha, N, P2O5 and K2O in the rainy and post-rainy seasons during the 8 years sequence cycles. The soil became enriched in organic carbon (OC) and maintained the available P while K concentration was reduced from medium to low status after 8 years. Integrated nutrient supply by incorporating glyricidia or FYM 15 days before transplanting the rice thereby substituting 25% nitrogen in the rainy season and application of only 75% of the recommended fertilizers in the post-rainy season maintained the overall mean yield at a level equal to that from the continuous application of the recommended fertilizer dose. The practice saved 30 kg N, 15 kg P2O5 and 10 kg K2O/ha in every season and improved the OC% and available P of the soil. The overall mean production of rice grain following the substitution of 50% nitrogen with glyricidia in the rainy season and fertilizer application at the recommended dose in the post-rainy season was also equal to that given by the application of the recommended fertilizer dose applied in every season and increased the OC% and available P of the soil. This procedure reduced the fertilizer dose in the rainy season by 60 kg N, 30 kg P2O5 and 20 kg K2O/ha.

Root re-growth, following aluminium (Al) stress, has been used as an indicator of Al stress tolerance. Genetic variation in root re-growth characteristics among eight triticale genotypes was investigated by a diallel analysis. Highly significant variation due to both general combining ability (GCA) effects and specific combining ability (SCA) effects indicated that both additive effects and non-additive effects were important in explaining the genetic variation for Al tolerance. The high estimates of heritability and the predictability ratio for root re-growth revealed the preponderance of additive genetic variance in the inheritance of Al tolerance. Differences in patterns of GCA effects and SCA effects among the parents provided strong evidence that the genetic control of variation for Al tolerance as assessed by root re-growth was a complex polygenic system. Three Al-tolerant genotypes, Tahara, Abacus, and 19th ITSN 70–4, were found to be the best general combiners for larger root re-growth, and they could be used in hybridization programmes to improve Al stress tolerance by following a simple pedigree method of selective breeding.

Herbage digestibility, which to a large extent governs the animal's intake, is not easily predicted for grazing animals, and can vary due to different grazing managements. The objective of this study was to build a model for a range of defoliation managements, which takes account of vegetative grass tiller structure, rather than dates of cutting or grazing, as done usually. An experiment was conducted in 1997, near Toulouse, to examine the digestibility of cocksfoot whole lamina and lamina segments (15 cm) for three defoliation regimes. These regimes mimicked severe and lenient intermittent defoliations plus a frequent defoliation treatment over a six-week period. Data were collected for (i) lamina mass and digestibility for the youngest expanded lamina, (ii) lamina appearance rate, lamina and sheath lengths.

Laminae digestibility, measured over time, varied greatly with the defoliation regime both for the youngest full expanded lamina and all laminae on a tiller. Digestibility of the youngest fully expanded lamina decreased from one leaf insertion level to the next, but this decrease was lowest when the defoliation regime was frequent. There was also a consistent decrease in the digestibility of a lamina from its tip to its base. Sheath and lamina lengths varied with treatment in a different way to lamina digestibility. During frequent defoliation, sheath and lamina lengths both increased slightly while they showed substantial increases in the two other treatments. Lamina length increased from one insertion level to the next, and resulted in the addition of less digestible lamina segments rather than reducing overall lamina digestibility. Sheath length, through its influence on leaf appearance rate and lamina length, could be used to predict the digestibility of laminae over time according to leaf insertion level and the difference in the digestibility observed from the tip to the base of a particular lamina. Sheath length could therefore provide a basis for predicting the time course of grass digestibility for different combinations of defoliation frequency and height.

We constructed a genetic linkage map of grasspea (Lathyrus sativus L.; 2n = 14) from 100 F2 individuals derived from a cross between PI 426891.1.3 and PI 283564c.3.2. A total of 71 RAPD, three isozyme and one morphological markers segregated in the F2 progeny. A small fraction of markers (12%) deviated significantly from the expected Mendelian ratio (1[ratio ]2[ratio ]1 or 3[ratio ]1). Out of 75 markers, 69 (one morphological, three isozyme and 65 RAPD markers) were assigned to 14 linkage groups comprising 898 cM. The average distance between two adjacent markers was 17·2 cM. The present linkage map will serve as a reference point for further linkage studies in grasspea.

The growth and yield of late-season tomato was studied at Akure, Nigeria, from 1995 to 1998. In each year, treatments were 0 (bare ground) or 12 kg/ha grass mulch applied to tomatoes at densities of 1·9, 2·8, 3·7 and 5·6 plants/m2 in a factorial combination. Increase in plant density and mulching reduced soil temperature at 5 cm depth while only the latter enhanced soil moisture at 10 cm depth throughout tomato growth. The improved soil hydrothermal regime resulted in earlier onset of flowering, 50% flowering date and date of first harvest respectively and shorter fruit harvest duration in tomato grown on either bare ground or at increased plant density.

Although increased plant density reduced shoot dry weight per plant at final harvest, both root length per plant and root dry weight per plant were enhanced. Fruit yield/ha increased as plant density increased although yield of the individual plants and their components were significantly reduced. Because of compensation at higher plant population densities, it is economical to grow late-season tomato at 5·6 plants/m2. Mulching enhanced growth and yield of tomato compared to bare ground and the result emphasizes the need for early and rapid growth of late-season tomato before the onset of terminal drought.

The efficacy of insecticides derived from two local plants was tested on termites infesting maize in maize–cassava–‘Egusi’ melon intercrops in two farmers' fields in south-western Nigeria during the 1989/90 and 1990/91 cropping seasons. The intercrops consisted of 40000, 20000, and 10000 maize plants/ha, with a fixed population of 10000 plants/ha of both cassava and melon. The insecticides were aqueous extracts of ripe seeds of neem (Azadirachta indica A. Juss) and the crude seed oil of Piper guineense Schum And Thonn, each at 10% concentration at the rate of 390 litres/ha and 18 litres/ha, respectively. On the plots treated with these natural insecticides, damage by termites [Microtermes spp., Macrotermes bellicosus Smeathman = nigeriensis Sjostedt and M. subhyalinus Rambur (Isoptera: Termitidae)] was significantly lower and maize yields higher than on the control plots. The mean percentage of maize stems attacked by termites (considering all categories of termite damage) in untreated plots during the trials ranged between 0–22% in 1989 and 2–46% in 1990. Planting densities did not significantly influence termite attack on maize. Cassava and melon plants or cassava tubers were not attacked by termites in any of the trials and their yields were not affected by insecticide treatments or planting density of maize.

An experiment was carried out in 1994 to examine energy and nitrogen utilization of lambs offered two contrasting grass-based diets. The two forages, which were from the same parent herbage, were grass silage and grass which was conserved by freezing. They were offered as sole diets or supplemented with either 250 or 500 g concentrates per kg total dry matter intake (DMI) to give a total of six experimental treatments. Seventy-two Dutch Texel × Greyface (Border Leicester × Blackface) lambs, consisting of 36 males which were initially 36 (S.D. 4·9) kg liveweight and 36 females which were initially 34 (S.D. 2·5) kg liveweight were used. Ensiling significantly increased apparent digestibility of dry matter, energy and nitrogen (P<0·001), but had no significant effect on methane energy loss as a proportion of gross energy intake, metabolizable energy intake (MEI), heat production, energy retained, efficiency of utilization of energy for growth (kg) or nitrogen retention. Supplementation of forage with concentrates resulted in a curvilinear decrease in heat production expressed as a proportion of MEI (P<0·05) and a linear increase in energy retention, expressed as an absolute value or as a proportion of MEI (P<0·05). Supplementation of forage tended to increase kg when calculated using Agricultural Research Council estimates of maintenance energy requirements, but had no significant effect when alternative estimates of maintenance were used. It is concluded that ensiling had no effect on efficiency of utilization of energy or nitrogen as measured by indirect calorimetry.

Dry matter intake is one of the major factors limiting milk production in dairy cows, although the quantity of food consumed by an individual cow when housed and fed as part of a group is rarely known. Such information would permit more precise ration formulation, concentrate allocation and selection of cows according to efficiency of milk production. Alkanes have been used with sheep and cattle to estimate feed intake under grazing conditions and could provide a technique for measuring intake in housed dairy cows. The purpose of this study was to examine alkanes C32 and C36, in combination with alkane C33, as indigestible markers for estimating intake of housed dairy cows fed on different diets under experimental and commercial conditions. Three experiments were conducted with hay-based diets (Expt 1), silage only diets (Expt 2) and a diet consisting of a silage-based basal ration plus concentrates (Expt 3). Animals were dosed once daily with C32 and C36, either on filter papers (Expts 1 and 2) or as part of a specially prepared concentrate (Expt 3). Faecal recoveries of alkanes ranged from 0·88 to 0·99. Over the range of intakes found in the three experiments (6–24 kg DM/d), the r2 values for estimated versus actual dry matter intakes ranged from 0·81 to 0·99. It is concluded that alkanes could provide a useful technique for estimating intake in dairy cows housed and fed in groups.

Field experiments were conducted at the farm of Qinghai Academy of Animal and Veterinary Science, Xining, China during 1996/97 to determine the effects of level of food intake on the urinary excretion of purine derivatives (PD), creatinine and nitrogen in yaks (Bos grunniens). Two experiments were carried out with three female yaks (initial body weight 173–187 kg, age 5 years). For Expt 1 a 3×3 Latin square experimental design was used with three levels of oat hay (nitrogen 13·5 g/kg dry matter (DM)) intake treatments, i.e. 0·3, 0·6 and 0·9 of voluntary intake (1·3–3·5 kg DM/d). Each treatment lasted for 17 days and the samples were collected during the last 7 days of each period. For Expt 2 the animals were fed the same oat hay as in Expt 1 for 3 weeks at a level equivalent to the estimated energy maintenance requirement (M) (1·5–2·2 kg DM/d). The intake was then reduced to 0·6 M on day 1, 0·3 M on day 2 and zero from day 3 until day 10. The animals were re-fed in the reverse order for 3 days. Of the PD, only allantoin and uric acid were present in the urine. The proportions of allantoin and uric acid were 0·86 and 0·14 respectively for both experiments. There was no response of creatinine and nitrogen excretions to feed intake. The rates of PD excretion per kg digestible organic matter (DOM) or digestible dry matter (DDM) were 13·5 and 13·6 mmol respectively. As expected, urinary PD excretion increased significantly (P<0·001) with increasing intake of DDM and DOM. The daily fasting PD, creatinine and nitrogen excretions amounted to 0·22±0·02 (S.E.), 0·25±0·01 mmol/kg W0·75 and 314±24·2 mg/kg W0·75 respectively. The results suggest that it is possible to establish a method for estimating intestinal microbial protein flow based on PD excretion in yaks.

The air quality in 82 German livestock buildings was investigated, using ammonia as a gaseous indicator. The ammonia concentrations were measured hourly and a 24 h mean was calculated. In cattle houses the mean ammonia measurement ranged between 3·7 ppm in calf houses (n = 16), 4·7 ppm in beef units (n = 10) and 6·4 ppm in dairy cattle houses (n = 8). In pig buildings the highest mean concentrations of 15·9 ppm were found in fattening pig livestocks (n = 8). In sow (n = 16) and weaner livestock buildings (n = 8) the averaged ammonia concentration varied between 13·4 ppm and 9·1 ppm, respectively. Within poultry houses, broiler flocks (n = 8) were associated with ammonia concentrations of 21·2 ppm. On the other hand, livestock buildings containing laying hens (n = 8) showed the lowest ammonia concentrations of all, namely 2·7 ppm. A significant positive correlation between temperature, relative humidity and ammonia concentrations could be found in poultry houses. Assuming an ammonia threshold limit value of 15 ppm, although no cattle house in this study exceeded this recommended limit, 31% of all the pig and poultry units investigated did, indicating a need for improved housing conditions in the future.

Two experiments were conducted, at ADAS Drayton in the autumn and winter 1996/1997, to compare methane (CH4) emissions from sheep housed either in a polytunnel system or in open-circuit respiration chambers. In each system, the sheep received maintenance levels of either cut grass or high temperature dried grass pellets (HTDG). All experiments in the tunnel were conducted on concrete to avoid any interactions of the CH4 with the soil/plant environment. The results suggested that CH4 production from the open-circuit chambers was greater than from the tunnel system (26·9±0·46 v. 31·7±0·35 l/kg dry matter intake (±S.E.) for open circuit respiration chambers and tunnel, respectively). Recovery tests gave similar results for both systems (95·5–97·9% for tunnels and 89·2–96·7% for chambers), and confirmed that both methods give good quantitative recovery of added CH4, and can therefore be assumed to provide reliable estimates of emissions from animals. There is no technical explanation, therefore, for the different estimates of emissions provided by the two systems. Further studies are required to understand the reasons for the differences and in particular, the possible links between animal behaviour induced by the two systems and CH4 emission rates.

Methods to alleviate the toxicity of Acacia angustissima (acc. no 15132) were evaluated in vitro and in vivo at ILRI Debre Zeit Research station in Ethiopia in 1997–1998. In an in vitro evaluation, A. angustissima leaves were incubated for 120 h with rumen fluid from Borana steers consuming native hay and cotton seed cake or from free ranging goat and sheep. Addition of polyethylene glycol (PEG, molecular weight (MW) 4000) increased gas production (4·9 v. 13·9 ml/100 mg dry matter (DM)), ammonia concentration (5·2 v. 9·7 mmol/l) and neutral detergent fibre (NDF) digestibility (380 v. 540 g/kg DM). In an in vivo experiment, Ethiopian highland sheep fed native hay and A. angustissima as a supplement (200 g/d) were inoculated with mixed rumen content from sheep gradually adapted on A. angustissima or were given PEG (MW 4000, 100 g/kg supplement) as methods to alleviate toxicity induced by A. angustissima. Sesbania sesban (acc. no 10865) with and without PEG were fed as a control. Effects on intake, apparent digestibility and nitrogen balance were studied in a 5×5 Latin square experiment. DM intake was increased by all the supplements (P<0·001) even though hay intake was slightly reduced compared to hay only (P<0·05). PEG increased intake of A. angustissima compared to inoculation. There were no significant differences in apparent DM or OM digestibility. Apparent nitrogen digestibility tended to be higher with S. sesban than with A. angustissima (P<0·10) and addition of PEG increased it with A. angustissima (P<0·05). Supplements increased ammonia and VFA concentrations and molar proportion of propionate (P<0·05). Supplements tended to increase microbial nitrogen supply (P<0·10) but there were no significant differences in efficiency of microbial synthesis. Supplements increased nitrogen retention (P<0·01) but due to short periods there were no significant differences in the live weights of sheep. All the animals remained healthy during the experiment suggesting that both methods used alleviated the toxicity of A. angustissima. Addition of PEG was more effective than inoculation in increasing intake and nutritive value of A. angustissima.

The Agricultural and Related Industries Study Group of the Operational Research Society held a one-day meeting on ‘Operational Research, Agriculture and the Environment’ on 27 November 1998 at Silsoe Research Institute, Bedfordshire, UK. The group promotes the use of the scientific method in solving management problems. The aim of this meeting was to look at models concerned with the effects of management of the environment in agriculture. Two looked at decision support systems – one for land use balancing different interests and one from the point of view of biodiversity. Andrew Moxey however highlighted the difficulties of interdisciplinary projects and the risk of the ‘illusion of technique’ in decision support systems. Three other papers were more specific – concerned with livestock and climate change, nitrogen fertilizer rates and slurry application.