Laboratory experiments were conducted in PVC columns to study the leaching and transformation of applied urea in sodic soils (Gharachon loam-Aquic Natrustalf and Domeli silty clay loam-Aquic Camborthid) reclaimed by gypsum application and kept submerged for 7 or 14 days after fertilizer application. The effect of different depths of irrigation water (5, 7·5, 10, 20 and 30 cm) on urea leaching was studied in a sandy loam sodic soil. In another experiment, the effect of time interval (0 or 4 days) between urea application and initiation of submergence with distilled water (for 7 or 14 days) was investigated involving two recently reclaimed sodic soils (Gharachon loam and Domeli silty clay loam). The results showed that the extent of urea leaching mainly depended upon soil texture. In Domeli silty clay loam, urea penetrated to 20 cm depth with peaks in concentration at 12·5 cm at both 7 and 14 days of submergence. In Gharachon loam urea-N moved to 25 cm depth after 7 days and to 35 cm after 14 days. In the sandy loam sodic soil peaks of urea-N concentration were observed at 12·5, 22·5 and 27·5 cm depths after infiltration of 5, 7·5 and 10 cm depth of water, respectively. Leaching with 20 and 30 cm depths of water moved urea deeper (below 50 and 70 cm, respectively). In recently reclaimed soils, leaching initiated immediately after fertilizer application displaced urea to slightly deeper layers compared with leaching initiated 4 days after urea application. Leaching may not be an important loss mechanism of urea-N in loam or silty clay loam sodic soils. However, in light-textured sandy loam sodic soils leaching beyond the root zone can be expected to create fertilizer management problems.

Mature Merino ewes (n = 500) were allotted at random to embryo removal (day 35; day 0 = day of oestrus detection), foetal removal (day 90) or lambing groups. These groups were further randomly divided into four single-sire mating groups. From day 1 until day 35 ewes were individually penned and fed either 50 or 150% of a maintenance ration (0·5 M or 1·5 M respectively). At day 35 single embryos were removed from anaesthetized ewes in the embryo removal group and all other ewes were endoscoped to confirm pregnancy. These ewes were then returned to pasture. Plasma samples were taken from all ewes on days 2, 4, 16, 23, 30 and 35 for analysis of glucose concentration. At day 90, ewes allotted to the foetal removal group were anaesthetized and plasma samples were obtained from the ewe's jugular vein and the umbilical arteryand vein. Ewes were then hysterectomized and the foetus was weighed and measured. Functional cotyledons were dissected from the uterus and chorio-allantois and all tissues were weighed. The remaining ewes (lambing group) were supervised at lambing and lambs were identified, weighed and measured.

Live-weight changes from weaning and wool production and quality were measured in ewe and ram lambs at their first shearing (11 months of age). Ovulation rates in the first two oestrous seasons of the ewes and wool production at their second shearing (2 years of age) were also measured.

During the 35-day treatment period, mean live-weight changes of ewes were –4·9 and + 1·8 kg in the 0·5 M and 1·5 M groups respectively. Pregnancy rates were similar in both groups but embryos from 0·5 M ewes weighed less than those from 1·5 M ewes (1·7 ± 0·04 ν. 1·9 ± 0·05 g; P < 0·005). Foetuses taken at 90 days from 0·5 M ewes were smaller than those from 1·5 M ewes but these differences reached significance (P < 0·025) only in the measurement of chin–crown length (8·0 ± 0·09 ν. 7·6 ± 0·11 cm). Correlations between foetal weight and total cotyledon weight, chorio-allantoic weight and empty uterine weight were all significant. Plasma glucose concentrations of ewes in the 0·5 M group were significantly (P < 0·001) reduced by day 9. Differences between the two nutrition groups in maternal and umbilical plasma glucose concentrations at day 90 were not significant, though foetuses from 0·5 M and 1·5 M ewes removed a mean of 30 and 11% respectively of available glucose from the umbilical vein. Differences in live weight between 1·5 M and 0·5 M animals at weaning and in the post-weaning period were not significant (P > 0·05). Wool production and woolquality were similar for both groups. Spontaneous ovulation rates measured on four occasions during the postpubertal oestrous season and on three occasions the following year were not significant (P > 0·05). Treatment with pregnant mare's serum gonadotrophin (PMSG) increased ovulation rates in the ewes of both groups; however, the differences failed to reach significance (P > 0·05); mean (± S.e.) ovulations per ewe were 4·8 ± 0·61 ν. 4·6 ± 0·60 for 1·5 M and 0·5 M group ewes respectively.

The influence of addition of sulphate on acetylene reduction in three paddy soils differing in their properties under two water regimes was investigated in a laboratory experiment. Nitrogenase activity was high in a P-deficient alkaline soil and addition of sulphate further enhanced the activity under two water regimes, with a pronounced stimulation under non-flooded conditions. Sulphate application to submerged alluvial soil enhanced nitrogenase activity with no apparent effect under non-flooded conditions. In acid sulphate saline Pokkali soil sulphate addition had little effect on the nitrogenase activity. Sulphate addition did not result in significant changes in the soil pH and redox potential. No relationship seemed to exist between the sulphate disappearance and stimulation of nitrogenase in these soils. A differential stimulation of N2-fixing microorganisms was noticed as a result of sulphate application. Results suggest that sulphate-induced stimulation of nitrogenase activity occurs in non-flooded soils.

Multifactorial experiments on winter barley cv. Igri grown after potatoes were made from 1981 to 1983 on silty clay loam soils at Rothamsted. All tested combinations of seven factors, each at two levels: with and without autumn pesticide (aldicarb), two sowing dates (September or October), with and without a fungicidal seed treatment (‘Baytan’), with and without spring and summer fungicides, two amounts of nitrogen, two times of applying nitrogen and with and without a growth regulator (‘Terpal’). Growth, development, yield, nitrogen uptake, pests and diseases were monitored. Sowing in September, fungicide sprays in spring and summer, and the growth regulator had the largest mean benefits on grain yield (+0·80, +0·56 and +0·34 t/ha respectively). Some factors interacted with sowing date; thus aldicarb, the fungicide sprays in spring and summer and the later timing of N all increased yield more on the September-than on the October-sown barley. The larger yields on the September-sown plots were associated with more ears/m2 (978 v. 744) and, in spite of fewer grains per ear (17·8 v. 20·1), more grains per m2 (17·6 v. 14·7 × 103), but lighter grains (39·2 v. 42·3 mg). The largest yields each year (ca. 8.0–8.5 t/ha) were obtained from September-sown barley fully protected from pests and from pathogens in spring and summer and given N in April rather than in March.

The aphid vectors of barley yellow dwarf virus were sufficiently common and infective in two of the three autumns to infect the September-sown barley sufficiently that their control by aldicar b enhanced yield. Nematodes, slugs and dipterous stem borers were not numerous enough to be damaging in any year. Mildew in autumn was controlled by the seed treatment, but effects on yield were inconsistent. Mildew in spring and summer was more abundant on the October-than on the September-sown barley; it was controlled by fungicide sprays, which increased yield significantly each year. Leaf blotch was more abundant on the September-sown barley.

Incidences of pollination and fertilization were surveyed in 21 commercial crops of field beans (16 of winter, 5 of spring) in southern Cambridgeshire (England) in 1982 and 1983. During the first part of the flowering season of winter beans, from 30 to 75% of flowers were not pollinated, and thus could not set pods. Where the pollination was largely spontaneous, relatively few pollinated flowers were fertilized and within them, relatively few ovules were fertilized. Later in the season and through most of the spring-bean flowering season, over 90% of flowers were pollinated and over 75% of ovules were fertilized. Incidences of pollination and fertilization varied greatly among the fields, and provision of hives of honeybees in fields of winter beans was associated with an improvement in the incidence of pollination from poor to moderate in 1983 but not in 1982. It is concluded that the incidence of pollination in many fields of winter beans is too low early in the season for an optimum distribution of yield along the stems.

Genotypes homozygous for specific PGI2 alleles were selected from four existing cultivars of different maturities, Gremie, Aberystwyth S.321, Barlenna and Melle. Each of three groups of plants (early BB selected from Gremie, intermediate CC selected from Aberystwyth S.321 and Barlenna and late AA selected from Melle) was isolated and allowed to interpollinate producing three experimental seed lots or populations. These populations, screened for their electrophoretic purity, were found to exhibit only the selected parental PGI2 genotype.

The three labelled populations were sown in boxes in September 1982 as monocultures, as all possible 50:50 mixtures, and as a three-way mixture to produce miniswards. Following one harvest in the establishment year these were harvested and dry-matter yields determined on a monthly basis from April to October 1983, two horizons in the sward being examined separately at harvests 1–5. Changes in the proportions of populations in each mixture were monitored by random subsampling of the cut herbage and subsequent electrophoretic testing.

The ability to dominate productivity in the top horizon in mixtures was dependent on the timing of reproductive development of any particular population. The patterns of productivity in the bottom horizon and in both horizons combined showed the early population dominating the mixed swards briefly at the beginning of the harvest but the late population dominating from June–July onwards. The intermediate population, although the most productive in monoculture, did not reveal any tendency to dominate productivity in any of the mixtures.

The potential use of electrophoretically labelled cultivars to study complex aspects of plant competition is discussed, along with the possible application of results from these studies to agricultural practice.

When given access to water once every 4 days, black Bedouin goats that were fed dry lucerne hay required, on average, only half the water and 60% of the gross energy they consumed when water was offered to them once daily. Digestibility of the feed in these goats was increased by 4%. The outflow of fluid from the rumen slowed down during the water deprivation period. A flow rate of 1080 ml/kg0.75. day was recorded during the 1st day of the period and only 350 ml/kg0.75. day during the last day. While being offered water once daily the outflow from the rumen amounted to 1243 ml/kg0.75.day.

Following the subjection of black Bedouin goats to infrequent drinking the mean retention time of the digesta in the rumen was extended by 33 % and in the entire gastrointestinal tract by 43 %. It is suggested that the slowing down of the ruminal flow that followed the subjection of the goats to infrequent drinking enabled the particulate matter to stay longer in the digestive tract despite their smaller size.

The withdrawal of food from Clun Forest ewes pregnant with twins produced signs of pregnancy toxaemia accompanied by a significant decrease in the concentration of glucose, a significant increase in the concentration of ketones, non-esterified fatty acids (NEFA) and 3-hydroxybutyrate in plasma and a significant decrease in total entry rate and irreversible loss of glucose. A single intramuscular injection of 0·05 or 0·2 mg/kg of triamcinolone acetonide had no significant effect on the concentration of glucose, ketones or 3-hydroxybutyrate in plasma or on the total entry, irreversible loss or recycling of glucose in ketotic or in normal pregnant sheep. The low dose had a significant effect on the concentration of NEFA in plasma. Recovery from the clinical signs was slow after either dose of steroid but appeared to be hastened by the onset of parturition which was more rapid after the higher dose.

The growth of 16 ram lambs was severely restricted for the first 6 weeks of post-natal life. Subsequently, these lambs (group R) were fed ad libitum. The diet was based on reconstituted cows' whole milk and lucerne chaff. A control (group C) of 16 similar lambs was fed ad libitum on the same diet from birth.

Lambs were weighed regularly and, in group C, four lambs were killed at the age of 1 day and then two at each of the following body weights: 10, 15, 20, 25, 30, 35 kg. In group R, five lambs were killed at the commencement of ad libitum feeding (age 43 days) and two each at the same body weights as in group C, except that only one lamb was available at 20 kg. After slaughter, the brain, liver, kidneys and the semitendinosus and gastrocnemius muscles were removed, weighed, stored and, with the exception of the liver, were analysed later for the following components: DNA, RNA and protein. Carcass weight and the weight of the kidney and channel (KC) fat were recorded. The femurs and metacarpals were removed from each carcass and cleaned, weighed and measured.

During the 6 weeks of restricted feeding, group R gained 0·9 kg while the ad libitum group C gained 13·5 kg. However, during recovery, group R grew faster than group C (0·37 ν 0·34 kg/day), reducing the weight for age difference near the end to 29 days at mean body weights of 30 kg.

Because of the design of the experiment, at the same age all measurements on group R animals, after the commencement of ad libitum feeding, were smaller than in group C. For this reason, the interpretation of the results has been based on differences between regression equations relating the various measurements to empty-body weight or to one another.

At the start of ad libitum feeding, brain weight, carcass weight, femur weight and femur length were bigger, while liver weight and KC fat weight were smaller in group R than in group C. At the end of the experiment, there were no significant differences between treatments for these measurements. Metacarpal shape differed between groups, the bone being relatively longer and narrower in group R than in group C, throughout the period of ad libitum feeding.

There were no significant differences between treatments in the relationships between DNA and corresponding tissue weights. However, the RNA was significantly less in both muscles in group R than in group C at the beginning of ad libitum feeding, but this difference had disappeared by the end of the experiment.

The brain protein: DNA and the brain weight: DNA ratios did not differ between treatments nor did they change significantly during the experiment. The semitendinosus was the only other tissue for which protein content was available and the protein: DNA ratio for this muscle differed between treatments, reflecting an acceleration of division by cell nuclei during the recovery period. The other tissue weight: DNA relationships did not differ between treatments and all ratios increased to values similar to those reported elsewhere. RNA:DNA ratios differed between treatments for both muscles, suggesting that high rates of protein synthesis occurred in group R during the recovery period.

In spite of the apparent normality of group R when measurements were related to EBW or tissue weights at the end of the experiment, only a long-term investigation would determine whether the weight-for-age difference of the type reported here would persist in adult life.

Two experiments were conducted, one with seven and the other with eight lactating cows, to measure the effect of supplementation with a cereal concentrate on the digestibility of fresh herbage and to provide equations relating digestibility of herbage to faecal indicators. Cattle were housed with free access to fresh herbage from individual Calan–Broadbent gates for recording feed intake, and faecal output was estimated by the use of chromic oxide. In both experiments a cereal supplement depressed herbage organic matter digestibility by 8 g/kg for each kg of fresh concentrate provided within the range 1–6 kg. The depression was associated with reduced digestibility of cellulose. Herbage digestibility was significantly related to faecal nitrogen and faecal cellulose. Herbage intake was depressed by concentrate supplementation and the response in milk yield to supplementation was small.

Field experiments conducted during wet and dry seasons with Ratna and Jaya varieties of rice (Oryza sativa Linn.) indicated that the use of Azolla pinnata as fertilizer increased growth in terms of number of tillers, leaf area index, dry-matter production, grain yield, N uptake and its accumulation in different plant parts. Azolla incorporation before planting in addition to 30 kg N/ha produced higher grain yields of both the rice varieties than 60 kg N / h a during the two seasons, while Azolla grown as an intercrop with rice plus 30 kg N/ha was superior to the 60 kg N treatment in grain production of Jaya during the wet season. Azolla incorporation before planting and also once intercropped with rice produced more grain yield than 60 kg N during the wet season. Among Azolla treatments, Azolla twice intercropped with rice produced comparatively lower grain yield. Azolla application increased the number of panicles/m2, number of grains/panicle and reduced sterility during both seasons. The N uptake of the rice crop in the treatments of Azolla incorporation before planting or intercropped once with rice in combination with 30 kg N/ha was similar to that of the 60 kg N/ha whereas the treatment of Azolla incorporation before planting plus once intercropped was similar during the wet season only.

Two experiments were conducted with four lactating cows in a latin-square design. The rations were composed mainly of maize silage plus concentrates, supplemented with urea, Pruteen, or soya-bean meal supplying 25% of total dietary N. Nitrogen and energy balances were measured, and milk production recorded.

Dry-matter digestibility, intake of metabolizable energy (ME) and ME concentration in dietary dry matter were not significantly altered when Pruteen was substituted for soya-bean meal. However, mean nitrogen digestibility for all the treatments was significantly lower than expected from published relationships.

The substitution of Pruteen for soya-bean meal did not significantly affect the nitrogen balance of the animals or milk production, but the latter tended to be higher than that recorded with urea.

The responses to the different N supplements are discussed in relation to the feeding systems of the Agricultural Research Council and the French Protéines Digestibles dans l'lntestin Grêle, and are compared with predictions made from statistical relationships developed in our laboratory.

In two experiments Dorset ewes were offered low (LP) or high (HP) herbage allowances during pregnancy to induce light (45–50 kg) or heavy (60–65 kg) post-partum body weights, respectively. During the first 6 weeks of lactation three herbage allowances were offered in a 2 x 3 factorial design: 2 (LL), 5 (ML) and 8 (HL) kg D.M./ewe per day. In the first experiment (Expt 1) lambs were removed and ewes machine-milked twice daily. In the second (Expt 2) they were suckled by two lambs; a further group on the ML allowance (MML) was machine-milked. Ewes were slaughtered during early pregnancy, post-partum and after 6 weeks of lactation for determination of body composition. Diet selection and faecal output were measured during lactation.

Absolute level of milk production was not affected by pregnancy nutrition when rearing two lambs and producing 2·0–2–7 kg milk/day or at 30–50% lower milk production when machine-milked. When expressed as g/kg W0·75 per day milk production was 14% greater in LP than in HP ewes. Milk fat content was higher at low allowances and was positively related to rate of body fat mobilization. Milk protein concentration was lower in suckled than in machine-milked ewes (40 ν. 47 g/kg) though total milk protein secreted (g/day) was similar. Milk production was 11–13% greater in ewes bearing twin lambs compared with those bearing singles whether machine-milked or suckling two lambs.

Mean intakes during the 6 weeks of lactation were 58, 69 and 73 g DOM/kg W0·75 per day, respectively, for LL, ML and HL groups when machine-milked and 51, 75 and 83 g DOM/kg W0·75/day, respectively, for ewes rearing two lambs. LP ewes consumed 13% (Expt 1) and 17% (Expt 2) more feed (g DOM/kg W0·75/day) during lactation than HP ewes, though in absolute terms (kg D.M./day) there was no difference. The difference in OMD between herbage offered (OMD = 0·77) and consumed ranged from 0 when sheep grazed to a residual herbage mass of 910 kg D.M./ha to + 0·04 when residual herbage was < 1600 kg D.M./ha.

All sheep, irrespective of pregnancy nutrition, milk production or herbage allowance, were in negative body energy balance during lactation. Mobilization of body fat during lactation averaged 40 g/day for machine-milked and 150 g/day for suckled ewes and was greatest in HPLL groups (92 and 219 g/day in Expts 1 and 2, respectively). In LP groups body protein loss during pregnancy was 10 and 16 g/day in Expts 1 and 2, respectively, and little further net protein loss occurred during lactation. The maximum rate of body protein loss in HP groups was 30 g/day in ewes rearing two lambs on the low herbage allowance.

Maintenance energy requirement during lactation averaged 0·238 MJ ME/kg W per day for ewes in both experiments; efficiency of use (Kl, (t)) of total energy available above maintenance for milk synthesis (namely, milk energy/ME intake - estimated energy requirement for maintenance (MEm + mobilized tissue energy) decreased from 0·84 to 0·51 with increasing rate of tissue energy mobilization. There was a positive relationship between Kl(t) and the proportion of mobilized energy derived from body protein.

The overall energy requirement of lactating ewes well-fed during pregnancy was similar to previous estimates (Agricultural Research Council, 1980) but that of sheep undernourished during pregnancy was 10–20% greater.

Water deficit during the panicle development stage reduced the grain yield of the main shoot panicle of pearl millet but this loss was compensated by increased grain yield of the tillers. The potential extent of compensation in grain yield components by tillers was investigated by removing the main shoot at panicle initiation (PI) and flowering stages respectively, for both irrigated and water-stressed plants. Grain yield loss by removal of the main shoot of plants at PI was fully compensated by tiller grain yield in both the irrigated and water-stressed plants. The compensation was, however, only partial when the main shoot was removed at flowering. The compensation for the grain yield loss in the main shoot due to either water stress or removal was through an increase in number of grains on the tillers. This increase was due to an increase in the number of productive tillers in the case of water stress and to both an increase in the number of productive tillers and an increase in the number of grains per panicle in the case of main shoot removal. This compensatory mechanism by tillers plays an important role in overcoming the effects of pre-flowering water stress damage to the main shoot.

The distribution of carboxymethylcellulase (CMCase) between various rumen fractions has been determined at three times after feeding in sheep containing no ciliate protozoa or five different protozoal populations. The total rumen CMCase was lowest in defaunated animals and tended to be higher in animals containing the amylolytic Entodinium caudatum or a natural mixed protozoal population and highest in sheep containing single cellulolytic species. In animals containing the cellulolytic species, Epidinium ecaudatum caudatum or Eudiplodinium maggii, 70% of the CMCase was associated with the protozoal fraction whereas less than 15% was present in a fraction prepared in the same way from sheep containing no protozoa or only Entodinium caudatum. The activity associated with the free bacteria was higher in sheep containing only Entodinium caudatum than in defaunated animals or those containing the cellulolytic species. Similar results were obtained 1·5, 6·5 or 24 h after feeding. Almost no activity was present in the cell-free rumen liquor under any condition. A variable amount of CMCase (up to 25% of the total) was associated with the plant debris in the rumen and the specific activity of the enzyme released by sonication was often higherthan that released from the free bacteria under the same conditions.

Differences in maternal body weight (12–15 kg) were induced in two groups (n = 43–50) of ewes during pregnancy by controlled allowance of pasture. Following parturition equal numbers (9–12) were allocated to one of three herbage allowances in a 2 x 3 factorial design during 42 days of lactation. This design was employed on two occasions; in one (Expt 1) the sheep were machine-milked twice daily, in the second (Expt 2) they each suckled two lambs. Balances of Ca and P in the body were determined by comparative slaughter during early pregnancy, at parturition and after 42 days of lactation. Herbage Ca intake and milk Ca output were determined on a weekly basis.

Mean daily milk production during the 42 days of lactation ranged from 1·48 to 2·68 kg/day amongst treatment groups. The general trend, irrespective of level of milk production, was for negative Ca balances, ranging from 0·71 to 1·42 g Ca/day. It is argued that variation in apparent absorbability (0·17–0·46) and in rate of absorption (30–113 mg/kg W per day) of Ca among groups suggest that these factors are not in themselves responsible for the poor Ca balances observed.

It is suggested that demand for Ca from the diet during early lactation is reduced as a consequence of bone Ca resorption due to matrix osteoporosis resulting from a protein deficiency state. Negative Ca balances in sheep undernourished during pregnancy were similarly attributed to poor protein balances.

Although negative P balances were observed these were not attributed to low P intake.

Stearoyl-CoA desaturase activity was determined in the subcutaneous and perirenal adipose tissue of pigs either fed ad libitum, or starved for up to 48 h. The effect of saturated fatty acids, as suet, or linoleic acid, as corn oil, on the enzyme activity was determined in pigs refed after starvation or fed 50% of the ad libitum intake.

Desaturase activity was similar in the outer and inner backfat and was twice as high as in perirenal adipose tissue. Starvation for 48 h decreased the activity by 70, 91 and 94 % in the outer and inner backfat and perirenal adipose tissue respectively. Lowering food intake to 50 % of normal decreased desaturase activity by 46% in the outer back-fat. There was no significant difference in the response of desaturase to 10% corn oil or suet added to the diet during refeeding after starvation or to the diet of underfed pigs. Although the enzyme thus appeared insensitive to dietary saturated fatty acids or linoleic acid, it was inhibited by linoleoyl-CoA in vitro.

The effects of different severities of continuous grazing imposed during the spring followed by a uniform continuous grazing management from midsummer onwards were examined in relation to sward morphology, herbage quality and the performance of young beef cattle in the years 1980 and 1981. The three grazing severities were based upon maintaining a constant sward height, namely 35 mm (severe), 50 mm (moderate) and 75 mm (lenient). From 1 June onwards a uniform grazing severity with a sward height of 60 mm was imposed. The grazing pressure on each paddock was adjusted by either adding or removing of cattle to maintain the target sward heights. When the grazing pressure was changed in June, the digestibility of both the herbage components on offer and the total herbage selected by the cattle was higher in the swards that had been severely grazed than that in the leniently grazed treatments. This appeared to be due to the combination of a higher proportion of younger, more digestible leaf laminae, less dead and less maturing true stem being present in the swards that had previously been severely grazed. Over the season as a whole, there was no significant difference between the grazing treatments in terms of individual animal performance or overall animal live-weight production per hectare.

Sheep were infused intraruminally with iso-osmotic solutions of saline, organic acids or lucerne silage extracts with or without additional acids and products of protein degradation found in silage. The post-ingestive effects of silage constituents on food intake were assessed by measuring the voluntary consumption of high dry matter silage over a 20 h period following feeding. In Expt 1, three different silage extracts each depressed food intake up to 4 h following feeding. In Expt 2, one of these extracts, to which 40 g of γ-amino butyric acid (GABA) was added, depressed intake up to 4 h following feeding; with the further addition cf 10 g α-amino butyric acid (AABA) and 28 g amines to this extract, intake was enhanced within 0·5 h but depressed from 4 to 8 h following feeding. In Expt 3, infusion of organic acids alone did not significantly depress food intake relative to saline, nor was there a discernible effect on food intake when graded levels of nitrogenous constituents from silage were added to the infusates. In Expt 4, infusion of an extract, to which only AABA and four other amines were added, depressed cumulative intake up to 8 h following feeding and had a more pronounced effect on food consumption than infusion of the extract with these constituents together with GABA. It was concluded that soluble constituents in silage can inhibit intake but no single constituent is primarily responsible.

The tannin content of milled seed of 55 lines of Vicia faba was evaluated and related to flower and testa colour, seed weight and percentage of testa. Tannin content in the lines analysed was related more to flower colour than to testa colour. No relationship existed between tannin content and seed weight. Lines with low tannin content had a low proportion of testa.