Nitrogen is an important factor determining grain yield of barley (Hordeum vulgare L.). Optimal rates and application timing can vary according to the environmental conditions under which the crop is grown. For 8 years (1979–86) barley yield, tillering, and leaf-area response to N application rates and timing were studied in southern Spain. The results showed that, under our experimental conditions, the most advantageous N rate for grain yield was 60 kg/ha, either split equally between sowing and tillering, or else with the greater proportion applied at tillering. Nitrogen applied at sowing increased number of tillers per plant, and N applied at tillering favoured tiller survival for later spike formation. In addition, N applied at tillering increased the number of leaves per plant, leaf area per plant, and the leaf area index, the maximum value of which occurred at anthesis.

Nine experiments were carried out on three different soil types in the south-east of Ireland over three years (1989, 1992 and 1993) to test the effect of early, normal and late sowing at three rates of fertilizer nitrogen (100, 125, 150 kg/ha) on the yield and grain quality ex-farm of spring malting barley (cv. Blenheim). Early sowing (January or February) was not possible in 1990 and 1991 due to wet weather. Sowing date had a pronounced effect on grain yield and grain N content in all three years. In 1989 and 1992, the earliest-sown barley (January or February) gave significantly greater yields than latersown crops (March and April) in five of the six experiments. The earliest-sown barley gave the lowest grain N in all six experiments and there was a gradual and significant increase in grain N content as sowing date was delayed. In 1993 the earliest-sown barley (February) gave significantly lower yield and greater grain N than the March-sown crop in all three experiments. The lower yield of the February-sown barley in 1993 was due to the significantly reduced number of grains/ear. The Aprilsown barley gave significantly lower yield and greater grain N than the earlier-sown crops in eight out of the nine experiments. In 1989 and 1992 only one of the six experiments, Ferns 1992, gave a significant yield response to increased rate of N (125 kg/ha). But in 1993, 125 kg N/ha significantly increased grain yield in all three experiments and 150 kg N/ha gave a further significant increase in yield in two of the experiments. Increasing increments of fertilizer N significantly increased grain N in all nine experiments.

The seasonal rate and extent of dry matter (DM) and neutral detergent fibre (NDF) degradation of the African sourveld grasses Alloteropsis semialata, Andropogon appendiculatus, Cynodon dactylon, Elionurus muticus, Eragrostis plana, Harpochloa falx, Heteropogon contortus, Hyparrhenia hirta, Microchloa caffra, Themeda triandra and Tristachya leucothrix in the rumen were measured using the nylon bag technique at the Döhne Agricultural Development Institute, South Africa, in 1993. The size of the soluble fraction (SF) was markedly different (P < 0·05) between species, being largest in A. semialata (22·6%) and smallest in E. plana (13%). Over all species, the mean SF was highest (P < 0·01) in spring (21·6%) and lowest in winter (11·8%). Species differed (P < 0·05) in the size of the potentially digestible fraction (PDF), with H. contortus and T. leucothrix having the largest (71%) and M. caffra the smallest (53%) PDF. Mean PDF declined (P < 0·01) from spring (77·3%) through to winter (55·8 %). The rate of DM degradation (kd) in the rumen also differed (P < 0·01) between species, with C. dactylon and A. semialata having the fastest (0·054) and E. plana and A. appendiculatus the slowest (0·039) degradation rates. Mean kd values were lower (P < 0·05) in winter (0·039) than in the other seasons (0·048).

Neutral detergent soluble (NDS) content was highest in A. semialata (29%) and lowest in E. plana (21·5%). Species similarly varied in the amount of potentially digestible cell wall (PDCW), with E. plana having the highest (57%) and C. dactylon the lowest (45·1%) PDCW content. Mean PDCW content was highest in spring (59·5 %) and lowest in winter (44·6%). Cell wall degradation (kc) rates were fastest (0·04) in C. dactylon and slowest in M. caffra (0·03). Overall, kc declined from spring (0·039) through to winter (0·029). In general, A. semialata, T. triandra and H. hirta appeared to have the most favourable, and E. plana, A. appendiculatusand M. caffra the least favourable, ruminal degradation characteristics. These differences suggest that species composition is likely to have a major impact on potential animal production on these low quality grasslands.

Seed production of annual medics was measured at four sites in Syria and Jordan between 1986/87 and 1988/89. Plots sown to medic in the first year were sown to wheat in the second year, and medic regeneration was measured in the third year. At three of the sites, adjacent plots were established in the second and third years and herbage production, and regeneration where appropriate, were measured in the third year. In this way, genotype × site interactions were used to assess medic adaptation to wheat/pasture rotations, continuous pasture and newly sown pasture.

Seed production was correlated with rainfall, the mean yield of 14 genotypes ranging from zero at 160 mm to 1 tonne/ha at 510 mm. Medicago polymorpha produced the highest yields, though not at the lowest rainfalls. Yields of M. rotata were also above average. There was, however, wide variation between species in seed weight, pod weight and seeds/pod.

Regeneration was better after wheat than after pasture. Herbage production was slightly better after pasture than after wheat, and both were greatly superior to that of newly sown pasture. After 2 years' germination, most genotypes retained more than half (mean 60%) and up to 99% of the seed produced in the first year.

Genotype × environment interactions were tested using analysis of variance (ANOVA), regression (Finlay & Wilkinson 1963) and principal component analysis (PCA). Because adaptation in pasture plants depends on many variables, PCA proved to be the most useful technique and the simplest to apply. ANOVA and regression were limited by their inability to process more than one variable at a time. Regression analysis is further limited by requiring a greater number of sites.

Based on seed production, seed survival, seedling regeneration and herbage production, certain locally collected genotypes of M. rigidula, M. noeana, M. polymorpha and possibly M. rotala, were adapted to conditions in Syria and Jordan. A similar methodology could be used to test pasture legumes throughout west Asia and north Africa

The leaf-curling plum aphid (Brachycaudus helichrysi Kalt.) is one of the main pests of sunflower (Helianthus annuus L.) in Europe. Yield losses were assessed in open field conditions and natural infestations over a 3-year period in Central West France. Plant variability and the choice of the more vigorous plants by aphids in open field conditions were two factors interfering with the experiments. To improve their sensitivity and to correct for biases without expanding the experimental design, an analysis of covariance was the best compromise. Field and cage experiments were conducted between 1985 and 1989 to assess the influence of aphids on the relationship between stem diameter at harvest and seed yield. As an indicator of plant vigour (size and yield), stem diameter at harvest was the best choice, as neither its relationship with seed yield nor its mean value was affected by aphid infestation. It was then used as a covariate to study yield losses. Aphid populations were quantified weekly from the beginning to the end of the infestations according to the leaf-curling symptom and to the number of aphids per plant. Multivariate analyses followed by cluster analysis identified groups of plants displaying similar infestation dynamics. It was found that when aphid populations reached > 100 per plant at the budding stage, yield losses occurred. When populations were < 100 aphids per plant at the budding stage and then decreased, no yield loss was observed. This gives growers ample time to monitor populations and treat crops.

Safflower (Carthamus tinctorius L.) is an important oilseed crop in India. A severe leaf spot disease caused by Ramularia carthami Zaprom. was observed in 1983 and investigations were carried out in order to analyse the role of environmental factors in disease development. High temperature and normal relative humidity, with dry rainless days, favoured disease development. Studies on the nature of in vivo resistance and susceptibility to R. carthami showed that growth of the fungal germ-tube was slower on resistant than on susceptible safflower cultivars, where the germ-tube extended with or without branching and penetrated infected leaves through the stomata. Of the four safflower cultivars evaluated, none were entirely free from the disease, but on the basis of lesion number and size, cultivar IGFRI-116 was rated as highly resistant.

Twelve accessions of pigeonpea (Cajanus cajan [L.] Millsp.) were grown on an Alfisol in a humid ecozone in Nigeria in 1989. At 4 and 8 months after planting (MAP), the primary growth was cut to determine the effect of age at first pruning on stump survival and coppice regrowth, and also on total herbage yield in a 12-month pruning cycle. It was clear that a height of 3 m or more is adequate for pigeonpea to receive its first cut with only 10% stool mortalities. Cutting the primary shoot at 8 MAP (in the dry season) ensured greater stump survival and also stimulated rapid coppice shoot regrowth, indicating the advantage of coppicing in the dry season and also at maturity. The total leaf and wood dry matter (DM) production in a 12-month cycle is greatest if the primary shoot was first cut at 8 MAP. This suggests that allowing the trees to become established is beneficial in the pruning management of pigeonpea. Accessions 11563 and 11575 have the ability to produce 7 t/ha of leaf DM from two prunings when first cut at 8 MAP, and significantly outperformed the local variety. The study revealed that pigeonpea does not coppice well and its biomass yields are much lower than that of Leucaena leucocephala.

Electrical conductivity was evaluated for estimating the nutrient content of cattle and pig slurries. Slurry samples were collected in 1991 from the storage tanks of 48 cattle and 10 pig units on commercial farms in Ireland. Samples were analysed for NH4+ and total concentrations of Na, K, Ca, Mg and P. Electrical conductivity (EC) was measured on raw slurries (ECraw) and on slurries diluted by 10 with water (ECdilute). Relationships between EC and nutrient content were examined by correlation and linear regression analyses.

In both slurry types, NH4+ was the dominant cation with K+ second in importance on a molar basis. Within each slurry type, the concentration of each of these cations was significantly correlated with EC. Using ECdilute gave more accurate predictions of concentrations than ECraw, but even ECraw was a better predictor than dry matter (DM) content. The linear relationships between NH4+ or K+ and ECdilute explained > 82% of the variance within each slurry type. The P content in slurries was related better to DM than to EC. Since EC measurement could be by cheap, robust meters, its potential for on-farm use deserves further investigation.

The fate of NO3- in raw and anaerobically-digested cattle slurries acidified with HNO3 was studied in a laboratory incubation experiment at the Agriculture and Food Science Centre, Belfast, in 1992. Four amounts (20, 30, 40 and 50 g) of each slurry type were acidified with five rates of HNO3 supplying 0·7, 1·4, 2·1, 2·8 and 3·5 mg N/g enriched to 10 atom% 15N. The slurries were stored in 500 ml glass jars for 7 days at 20 °C. The jars were left open to allow O2 supply by diffusion through the slurry surface to create varying degrees of aeration depending on slurry depth. Fluxes of N2 and N2O were measured from each jar for 2 h each day. The pH was measured after each gas flux measurement and HNO3--N recovery as NO3--N measured at the end of the incubation.

For raw cattle slurry, almost none of the HNO3--N was recovered as NO3--N at the 0·7 or 1·4 mg HNO3-N/g rates, but recovery was not significantly different (P > 0·05) from 100% at the 2·8 and 3·5 mg HNO3-N/g rates. At the 21 mg HNO3-N/g rate, recovery was higher when larger amounts of slurry were incubated, indicating that O2 could promote NO3-loss. Average pH during the incubation period had to be < 5·5 for recovery of HNO3-N as NO3-N to be > 90%. For digested cattle slurry, recovery was > 87% at the 0·7 mg HNO3-N/g rate and not significantly different (P > 0·05) from 100% at all other acid rates.

Fluxes of N2 and N2O were not detected from any digested slurry treatment but were measured in all raw slurry treatments. Gaseous losses occurred after incubation for 1 day, and maximum flux rates occurred after incubation for 2 days. Fluxes were highest from incubations with the intermediate amounts of slurry containing 1·4 mg HNO3-N/g. The largest observed flux from an individual jar was 322 μg N2-N/g/2 h with 40 g of slurry containing 1·4 mg HNO3-N/g, which would have accounted for a loss of 23% of the added HNO3-N during this period. Nitrous oxide and N2 were always evolved simultaneously and when the pH was between 7 and 9. On average, 90% of the total loss of N-gases was as N2.

The raw slurry contained much higher concentrations of volatile fatty acids than the digested slurry. If HNO3 is added to slurries containing such readily decomposable substrates, the NO3- is susceptible to loss as N2 and N2O by denitrification. Oxygen supply, pH and substrate availability control the rate of denitrification.

Field experiments were carried out in Brittany (Western France) in 1993 to measure ammonia losses from surface-applied pig and cattle slurry. Experiments were conducted on grass, stubble (wheat and maize) and arable land using a wind tunnel system. Ammonia losses were followed during periods ranging from 20 to 96 h after slurry application. Rates of slurry applied varied from 40 to 200 m3/ha. In two experiments, losses from cattle slurry were respectively 75 and 54% of the ammoniacal nitrogen applied in the slurry. Ammonia emissions from pig slurry applied at a rate of 40 m3/ha, during spring and summer experiments, were higher on grass (45–63% of the total ammoniacal nitrogen applied) than on wheat stubble (37–45%). On average, 75% of the total loss in all experiments occurred within the first 15 h after spreading. Significant correlations were found between ammonia losses (kg N/ha) and mean soil temperature and slurry dry matter content (%) using simple linear regressions and stepwise procedures. The time of application was also found to influence the magnitude of ammonia loss: 83% of the total loss occurred within 6 h when the slurry was applied at midday compared with 42% when it was applied in the evening.

Rape (Brassica campestris) was labelled, during cultivation, with 15N and then prepared into rapeseed meal (RSM) and rapeseed straw (RSS). Three non-lactating cows, equipped with ruminal and Tshaped duodenal cannulae, were used to incubate both feeds in the rumen for various periods of time in mobile bags contained within larger mesh bags. Some of the mobile bags were then inserted into the duodenum and collected from the faeces. The effects of rumen incubation time and contamination by non-feed N during the rumen incubation and subsequent intestinal digestion on the intestinal feed N digestion values so obtained were assessed. The intestinal digestion of intact RSM nitrogen was also measured in vivo by dosing directly into the duodenum and measured as the net loss of excess 15N not recovered in the faeces.

With RSM, microbial contamination did not affect the disappearance values of feed N from mobile bags in the rumen, the difference in the disappearance between 15N and total N (NT) being negligible. However, the decrease in the disappearance from the mobile bags in the intestine with increasing rumen incubation time tended to be smaller with 15N than with NT, resulting in significantly increasing differences between 15N and NT with increasing rumen incubation time. The average in vivo net loss within the intestine of 15N of labelled RSM was much lower (50·0%) than the 15N disappearance from the mobile bags (80·9%).

With RSS, the disappearance values of 15N and NT from the mobile bags in the rumen were different. The disappearance of 15N was always higher than that of NT, and it also increased significantly more with increasing incubation time. The disappearance of 15N from the mobile bags within the intestine decreased with increasing rumen incubation time, but that of NT remained rather constant.

It is concluded that the intestinal N digestion of a feed may vary significantly depending on the degree of preceding rumen degradation. Therefore, constant intestinal digestion coefficients may not be applied in the modern protein evaluation systems. With protein-rich feeds, such as RSM, the influence of non-feed N is small and does not interfere with the estimation of intestinal N digestibility by the mobile-bag method. With fibrous feeds with low N, such as RSS, intestinal N digestion values are seriously confounded by non-feed N. The present study with RSM did not confirm the correspondence between the value obtained by the mobile-bag method and that in vivo.

Two studies were conducted between 1991 and 1993 to assess the impact of grazing improved v. natural pasture at different stocking densities on animal and vegetation responses. In the first grazing season, goats of the West African Dwarf (WAD) breed were grazed at four stocking densities – 1·90, 2·86, 3·86 or 4·71 TLU/ha (Tropical Livestock Unit = 250 kg liveweight). Wet-season grazing lasted for 14 weeks while assessment was continued until kids were weaned at 120 days of age. There were significant (P < 0·05) changes in the content of legume and forbs in response to variations in stocking density. Liveweight gain per head was significantly (P < 0·05) higher at low than at high stocking density. The same trend was maintained with most aspects of reproductive performance. Goats at low stocking density were of a higher (P < 0·05) body condition score at parturition and had larger litter sizes than does stocked at high stocking density. Kid birth weight and liveweight gain were significantly (P < 0·001) higher at low than at high stocking density. The overall weight loss in does during nursing was similar at the different stocking densities.

In the following year, comparisons were made between three stocking densities, 2·11, 3·17 and 4·23 TLU/ha, on improved and unimproved pastures. Over 14 weeks of wet-season grazing, there was a significant (P < 0·01) decline in available dry matter except at the lowest stocking density on both pasture types. Weight gain per head decreased with increase in stocking density while the reverse was true for weight gain per unit of land. No significant differences were observed for reproductive performance between pasture types or stocking densities. Kid growth rate up to weaning declined with increase in stocking density (P < 0·001) while kid mortality rose with increase in stocking density (P < 0·01). Weight losses in nursing does were higher (P < 0·05) at low than at high stocking density.

The voluntary food intake (VFI) and apparent digestibility of chopped timothy hay by six Scottish Blackface (BF), Shetland (SH) and Dorset Horn (DH) non-breeding ewes were assessed in summer and winter. In two experiments VFI was higher in summer than in winter but did not differ between the breeds. In Expt 1, the seasonal difference in VFI was greater in BF and SH than in DH ewes and the apparent digestibility of dry matter (DM), organic matter (OM) and total nitrogen was consistently higher in all three breeds in summer than in winter. The higher digestibility of DM of the hay in summer than in winter was confirmed in Expt 2, in which the whole tract mean retention time (MRT) of the undigested residues was measured using 103ruthenium-phenanthroline as a marker. The MRT was lower in all three breeds in summer than in winter. In a third experiment, following at least 6 months ad libitum feeding on the hay, for all three breeds the contents of DM and liquid in the reticulorumen were smaller in summer, and the liquid fraction formed a greater proportion in summer than in winter. The results contrast with the expectation that intake and digestibility within a given forage type are negatively correlated. Although an increase in digestibility may facilitate a higher VFI in summer (BF and SH) it does not necessarily do so (DH).

Two grazing experiments were undertaken to evaluate the effects of sulla (Hedysarum coronarium) which contains condensed tannins (CT) and lucerne (Medicago sativa), which does not contain CT, on the performance of parasitized and non-parasitized lambs. The work was carried out near Palmerston North, New Zealand in 1992 and 1993.

Experiment 1 used lambs which were either genetically ‘conventional’ or ‘resistant’ to gastrointestinal nematodes and which had a naturally acquired heavy parasite burden (faecal egg count (FEC) 1300–2500 eggs per gram faeces (epg)). They were either drenched to remove all the parasites or remained undrenched when allocated to pure swards of either sulla or lucerne for the 28-day experimental period. Liveweight gains of drenched lambs were higher with sulla (mean 302 g/day) than with lucerne (mean 245 g/day; P < 0·02) and rates of gain were not reduced (P = 0·10) by genetic resistance to nematodes. Undrenched lambs had mean faecal egg counts of 1090–2220 epg after 28 days, with higher counts in lambs grazing lucerne than in those grazing sulla (P < 0·05). Lambs grazing lucerne also had a much lower average daily gain (ADG) (mean 50 g/day) than those grazing sulla (mean 206 g/day; P < 0·0001). Resistant lambs had reduced (P = 0·07) ADG in the sulla treatment only.

In Expt 2, lambs with a relatively low worm burden (average FEC 250 epg) were either drenched to remove all worms or dosed with 20 000 Trichostrongylus colubriformis infective larvae and grazed on pure sulla or lucerne swards for 42 days. Drenched lambs had a similar liveweight gain (mean 190 g/day) and wool growth when grazing either sulla or lucerne. Undrenched lambs grazing sulla had a greater ADG (mean 129 g/day) than those grazing lucerne (mean −39 g/day; P < 0·0001) and a higher rate of wool growth from mid-side patches (P < 0·009) and larger wool fibre diameter (P < 0·05) than those grazing lucerne. Undrenched lambs grazing sulla had lower FECs (P < 0·05) on days 14, 21, 35 and 42 than those grazing lucerne and lower (P < 0·05) Trichostrongylus colubriformis worm burdens at slaughter. Undrenched lambs had less perineal faecal contamination when grazing sulla rather than lucerne. Parasite-induced anorexia was evident in the lambs grazing lucerne but not in those grazing sulla.

These findings demonstrate that the performance of lambs infected with nematode parasites can vary substantially when given feeds of a similar chemical composition. The superior performance of lambs grazing sulla was probably caused by condensed tannins, which decrease protein degradation in the rumen and can increase post-ruminal protein availability.

A feeding trial was undertaken in Palmerston North, New Zealand in 1991, to determine the extent to which condensed tannins (CT) in Lotus pedunculatus were able to affect the nutritive value of ryegrass (which does not contain CT) for sheep. The trial involved three groups of 11 male cryptorchid sheep held in metabolism crates so that intakes, digestibilities and nitrogen balances could be measured over a 42-day feeding period. One group was fed freshly cut ryegrass (Lolium perenne L.) as a sole diet (Grass group) and the other two groups were given a mixture of about one third freshly cut lotus and two thirds ryegrass (dry matter (DM) basis). This mixture contained c. 1·8% CT in the DM. One group given the mixed diet was drenched twice daily with polyethylene glycol (PEG; PEG group) to bind and remove the effects of the CT from the diet, whilst those not drenched were designated the ‘Tannin’ group. Intakes of all groups were constrained to that of the Grass group of sheep.

Condensed tannins in the lotus had a major effect on the digestion of ryegrass. When the forages were immature, apparent digestibility of nitrogen (N) was substantially reduced in the Tannin sheep (65·3%)compared to the PEG (77·5%) and Grass (77·9%) groups (P < 0·001). With mature forages, the respective values were 48·8, 62·5 and 53·7% (P < 0·01). Rumen ammonia concentrations and plasma urea concentrations were also reduced by CT. Condensed tannins reduced DM digestibility by 3·7 percentage units (P < 0·05) mainly through the effects on N digestion. Concentrations of acetate, propionate and butyrate in rumen liquor were not affected by treatment, but CT reduced the proportions of the minor volatile fatty acids (VFA; isobutyrate, valerate and isovalerate). Treatment did not affect liveweight gain (131—164 g/day), or wool growth from mid-side patches, and N retention was not reduced by CT, so that the nutritive value of the forage was maintained despite the reduction in N digestibility. The principal finding was that CT in Lotus pedunculatus were able to affect the digestibility of both grass and lotus when fed together, and that as little as 1·8% of CT in the dietary DM had a substantial effect on rumen function.

Degradation of biogenic amines was studied in rumen contents obtained from wether sheep adapted to diets with different levels of biogenic amines: high (H), low (L) and without (W), containing 7·4, 2·4 and 0 g amines/kg dry matter (DM), respectively. To 200 g of the rumen contents (RC), 2 ml of a solution containing a mixture of the biogenic amines: cadaverine (73·5 mmol/1), histamine (45·0 mmol/1), putrescine (830 mmol/1) and tyramine (123·5 mmol/1) were added, followed by a 5 h incubation in vitro. The fermentation pattern in RC derived from H and L differed from that in RC derived from W. This difference was attributed to differences in fermentative properties of silage and hay-based diets in the rumen. The addition of amines increased ammonia production, which was highest in RC from sheep adapted to silage with the highest amine content (diet H). Amines had no influence on gas production. Amine degradation occurred in all types of RC, but the extent depended on adaptation of the rumen microflora, such that 709, 54·2 and 25·3% of the added quantity in RC from H, L and W, respectively, was degraded. Generally, the breakdown of the individual amines was highest for histamine, followed by tyramine, putrescine and cadaverine. Tyramine breakdown was particularly slow in RC from diet W. These results imply that in animals adapted to grass silage with high concentrations of biogenic amines, the accumulation of amines in the rumen will be prevented by an increase in the amine-degrading capacity of the rumen microbes.

Two experiments done at the University of Ilorin, Nigeria, in 1992/93 investigated the effects of dietary levels of thevetia cake on the performance and nutrient retention of broilers. In Expt 1, dayold broiler chicks were fed either a control diet or diets containing 5, 10 or 15% thevetia cake during a 4-week period. In Expt 2, day-old broiler chicks were fed a standard diet until 4 weeks of age and were then switched to diets containing 0, 5, 10 or 15% thevetia cake. Inclusion of thevetia cake in broiler diets, irrespective of level of inclusion, drastically reduced feed intake and weight gain (p < 0·01) at both the starter and finisher stages. Dietary thevetia cake also caused a significant increase in mortality rate (P < 0·01). There was a reduction in protein and fibre retention and dietary ME with increases in the dietary level of thevetia cake at both the starter and finisher stages (P < 0·05). It is concluded that thevetia cake is toxic to broilers and needs further processing before it can effectively be used as an ingredient in broiler feed.