The effects of exchangeable sodium on emergence, growth, development and yield composition of oilseed rape were investigated at Buenos Aires, Argentina in 1992. A pot experiment was performed using five exchangeable sodium levels, expressed as Sodium Adsorption Ratio (SAR: 12, 20, 27, 34 and 44). Soil with nine different exchangeable sodium contents (SAR from 0·5 to 50) and seven different particle size distributions (from sandy loam to clay) were used in a parallel trial to study the effect of soil crusting on oilseed rape emergence. Soil crusts were made using a rainfall simulator after seeds were sown. Both experiments showed that the direct effect of sodium on emergence occurred with SAR values higher than those which caused clay dispersion (SAR > 20). Oilseed rape seedlings could penetrate crusts having a resistance of < 230 kPa. At SAR values > 20, main stem growth and yield decreased significantly (P < 0·05). These reductions were counterbalanced by an increase in the number of secondary stems. At SAR levels > 34, stem number, grain number per pod on the main stem, as well as pod number on both main stem and secondary stems, were reduced, leading to a marked reduction in total yield. The most important agronomic effect of soil sodium on oilseed rape would be at emergence stage, due to soil crusting.

Lucerne (Medicago sativa L.) sowing rates were evaluated to determine how environmental conditions affect seedling establishment and plant survival. Lucerne (cv. Peace) was sown during 1988 at rates ranging from 2·8 to 11·2 kg ha-1 at one irrigated and two dryland sites (a forest and a mid-elevation grassland site) near Kamloops, British Columbia, Canada. Dry matter (DM) yield was measured for 3 years, and plants were counted to estimate seedling establishment and plant survival. Sowing year DM yield increased linearly with sowing rate at the irrigated and forest sites, but at the mid-elevation grassland site it decreased at sowing rates > 5·6 kg ha-1. In the first full-production year, DM yield levelled off at sowing rates > 8·4 kg ha-1 at the irrigated site and was not affected by sowing rate at the forest site. At the grassland site, first full-production year DM yield decreased at sowing rates > 2·8 kg ha-1. In the second full-production year, there was a linear effect of sowing rate on DM yield at the irrigated site, but it did not affect DM yield at the two dryland sites. At all sites, the percentage of seedlings that survived to the end of the second growing season decreased curvilinearly as seedling density increased. Survival was less at the irrigated site than at the dryland sites. Higher sowing rates are required at irrigated sites than at dryland sites to obtain maximum yield or a required plant population density.

Potato seed tubers inoculated with Rhizoctonia solani and Erwinia carotovora subsp. atroseptica, to induce stem canker and blackleg respectively, were planted with uninoculated seed tubers in field experiments designed to measure the effects of the diseases and of neighbouring plants on tuber yield. Gaps were also included. The plant variables total and ware (> 150 g) yields and tuber numbers were affected by disease, and also by competition from the two plants on either side in the same row (first neighbours), and increased as competition from neighbouring plants decreased. Plants adjacent to the first neighbours also influenced yields of plants with stem canker but those in adjacent rows did not have a significant effect with either disease. Both diseases altered the tuber size distributions, which were also modified by neighbouring plants. The data were used to predict total and ware yields for crops containing different proportions of healthy, diseased and missing plants.

Field experiments were conducted to test the seed yield responses of winter oilseed rape (Brassica napus L., cvs Libravo or Falcon) to the addition of different rates of S fertilizer, at three N application rates, on a sandy loam at Woburn, Bedfordshire, in 1990/91, 1991/92 and 1993/94. Large increases in seed yields, ranging from 0·7 to 1·6 t/ha, or 42–267% on a relative scale, were obtained in response to the application of 40 kg S/ha with 180 and 230 kg N/ha treatments. The effects of S were highly significant in 1991/92 (P < 0·01) and 1993/94 (P < 0·001) and close to significant (P = 0·053) in 1990/91. The yield benefits were obtained mainly from the application of the first 10 kg S/ha and further yield increases were unlikely above 40 kg S/ha. Increasing N application from 180 to 230 kg/ha decreased seed yield in 1990/91 and 1993/94, when no S was applied. In contrast, seed yield was not increased by S at zero or low (50 or 100 kg/ha) N rates. The interactions between N and S on seed yield were significant (P < 0·05) in 1990/91 but not in the other two seasons. Application of S also increased seed oil content in 1993/94, when the degree of S deficiency was particularly severe. With an application of 230 kg N/ha, the crops took up 5–22 kg S/ha at maturity when no S was applied and 26–51 kg S/ha when 40 kg S/ha was applied. The utilization efficiency of the fertilizer S ranged from 50 to 73% in the three seasons. Although the concentrations of total N in plants were largely unaffected by S treatments, large amounts of NO3-N accumulated in the leaves of S-deficient plants in 1993/94. This indicates that N metabolism was disrupted by S deficiency. The concentrations of S and the N: S ratios in different tissues and the whole plant changed considerably with time. The concentration of S in leaves at early flowering was found to be the best index in predicting S deficiency in terms of seed yield, and a critical value of 3·8 mg/g was obtained. In comparison, the N: S ratio in leaves at early flowering was a much poorer predictor of S deficiency.

The recombinant genotypes generated with a potato breeding scheme cannot always be evaluated over the range of environments for which cultivar development is aimed. At the Scottish Crop Research Institute, the cultivar breeding programme aims to develop genotypes which perform well in one or more environments in the UK and/or countries around the Mediterranean basin. In each of three years (1984, 1985 and 1986) between 40 and 50 potato genotypes were selected solely on their phenotypic performance in an environment just south of Edinburgh in Scotland. These were then trialled at sites in four environments in the typical ware-producing regions of England and four environments around the Mediterranean region. Despite significant genotype x environment interactions for many traits, the superior genotypes tended to perform well over a wide range of environments.

Aspects of the photoregulation of phenological development in bambara groundnut were studied in a glasshouse experiment in The Netherlands. The influence of a 14 h photoperiod (which retards podding) during a period of variable length prior to an 11 h photoperiod (which induces podding) on flowering, yield and on the position of pods on the plants was determined. The third generation of three plants of genotype ‘Ankpa 4’ from Nigeria was used as the split-plot factor in a split-plot design with three replicates. The main plots were four daylength treatments: a period of 21, 28, 41 or 54 days under the 14 h photoperiod before transference to the 11 h photoperiod. Plants transferred after 28 or more days started flowering sooner the earlier they were transferred. Plants transferred after 21 and 28 days began flowering at the same time (51 days after sowing), thus indicating juvenility. At harvest, 135 days after sowing, the total seed dry weight per plant was higher for plants transferred after 41 and 54 days than for plants transferred after 21 days. Plants transferred after 28 days gave an intermediate value. Most (79–91%) pods were produced on branches that developed on nodes 1–4 of the main axis. There were no differences in the fractional distribution of the pods along the main axis in plants transferred after 28 or more days. Pods of plants transferred after 21 and 28 days were more evenly distributed over the nodes of the first two primary branches than pods of plants transferred after 41 and 54 days. Plants of the latter treatments produced their pods more towards the tips of the branches and concentrated on two neighbouring nodes (nodes 5 and 6 for plants transferred after 41 days and nodes 6 and 7 for plants transferred after 54 days). Delaying the induction of podding in this experiment therefore resulted in higher seed yields per plant and a more synchronized development and hence maturity of pods.

Soil mineral nitrogen (Nmin) was measured to 90 cm at a total of 12 sites in the UK in the autumn after an oilseed rape experiment, which measured responses to fertilizer N. On average, Nmin, increased by 15 kg/ha per 100 kg/ha fertilizer nitrogen (N) applied to the rape, up to the economic optimum amount of N (Nmin). There were larger increases in Nmin where fertilizer applications exceeded Nopt, thus super-optimal fertilizer applications disproportionately increased the amount of nitrate likely to leach over-winter. The small effects of sub-optimal N on Nmin were associated with large increases in N offtake by the oilseed rape, whereas the larger effects of super-optimal N on Nmin were associated with only small increases in N offtake. Over 70% of the variation in autumn Nmin was explained by the previous rape's N fertilizer rate and the topsoil organic matter content.

Nitrogen applied to the rape increased grain yields of the succeeding wheat crops when no further fertilizer N was applied to the wheat. It was concluded that N applied to oilseed rape significantly affected Nmin after harvest, and these effects were not completely nullified by leaching over-winter, so soil N supply to the succeeding wheat crop was significantly increased. Responses in grain yield indicated that each 100 kg/ha N applied to the rape provided N equivalent to c. 30 kg/ha for the following cereal. Each 1% of soil organic matter further contributed N to the wheat, equivalent to 25 kg/ha.

It is important to ensure that oilseed rape receives no more than the optimum amount of fertilizer N if subsequent leaching is to be minimized. Reductions below optimum amounts will have only a small effect on leaching. Substantial changes in the economic optimum N for rape production should be accompanied by adjustment in fertilizer N application to following wheat crops. Fertilizer recommendation systems for wheat should take account of the fertilizer N applied to the preceding oilseed rape and the topsoil organic matter content.

Stands of bambara groundnut (Vigna subterranea (L.) Verde.) were grown in five controlledenvironment glasshouses at the Tropical Crops Research Unit, University of Nottingham, Sutton Bonington Campus, in 1990. Five soil moisture regimes were imposed (one per house), from fully irrigated each week (treatment A), to no irrigation after crop establishment at 35 days after sowing (DAS) (treatment E). Decreasing the amount of water applied resulted in a decline in total dry matter production and harvest index, and a reduction in pod yield from 412 (treatment B) to 0·041 ha-1 (treatment E) at 125 DAS. A maximum leaf area index of 5–4 was achieved by treatments B and C at 90 DAS, resulting in a fractional interception of c. 0·8 of incoming radiation. Total accumulated radiation interception values were 749, 693, 688, 618 and 554 MJ m-2 for treatments A, B, C, D and E, respectively. The efficiency of conversion of the radiation intercepted into dry matter was reduced from 1·41 to 0·50 g MJ-1 by drought.

The biology of some host-adapted populations of A. gossypii and the comigration band mobilities of aphid enzyme extracts were studied in 1992 in Kochi, Japan. Results indicated that biologically there are no significant differences between eggplant, cucumber and watermelon aphid populations with regards to longevity and the number of offspring produced per female aphid. However, in all populations, adult survival declined significantly within 5 days of introduction onto a new host from the original. No genetic differences were observed between populations since they scored equally at seven different enzyme loci: malic enzyme (ME), glucosephosphate isomerase (GPI), isocitrate dehydrogenase (1DH), phosphoglucomutase (PGM), 6-phosphogluconate dehydrogenase (6-PGD), malate dehydrogenase (MDH) and naphthylacetate linked esterase (EST). Also, protein bands, as revealed by SDS-PAGE, as well as carboxylesterase banding patterns, showed no differences between populations. These results indicate that the various host-adapted populations of A. gossypii infesting a wide range of solanaceous and curcubitaceous crops can be treated as single species devoid of genetic variability. The slight differences in biology and behaviour found between the various hostadapted populations can be attributed to physiological variations, differences in host nutrient composition, and possibly to some unknown co-existing biotic factors.

The effects of overwinter cover cropping, delayed ploughing and method of straw disposal on the quantities of nitrate leached (averaged over three winters during 1989–93) from a chalk loam in Eastern England were examined. The recovery of ‘retained’ nitrogen (retained through cover crop uptake, delayed ploughing and immobilization by straw) in a following spring crop was also assessed. In the first two winters, the rye cover crop decreased nitrate leaching by > 90% (28 kg N/ha per year), as compared with bare fallow treatments. In 1992/93 this decrease was only 23% (10 kg/ha), due to the early onset of drainage before cover was well established. Delayed ploughing on bare treatments, to decrease autumn N mineralization and subsequent nitrate leaching, was ineffectual in 1989/90 but had substantial effects in 1990/91 and 1992/93; N mineralization, inferred from soil mineral nitrogen content, and nitrate leaching were decreased by 31 and 35% in 1990/91 and by 36 and 61% in 1992/93, respectively. Nitrate leaching (averaged over three winters) was unaffected by straw incorporation. There was no evidence of recovery of cover crop N in the spring sown test crops (barley or sugarbeet). In the low soil N input situation encountered in this experiment, it was unnecessary to sow cover crops before early September in years of average or below average rainfall to ensure that the average soil solution concentrations remained below the EU drinking water limit of 11 mg NO3-N/1. However, in wetter seasons substantial N leaching occurred before cover had taken up much N. In 1992/93 N retained against leaching by a rye cover crop in previous years was apparently being remobilized and lost through leaching, although if cover was grown again there was less leaching than from bare land. In the future, an increase in the extent of cover cropping might increase transpiration rates and therefore lead to a decrease in aquifer recharge.

The response of four cowpea (Vigna unguiculata(L.) Walp.) cultivars to the warm, semi-arid tropical environment at the ICRISAT Sahelian Center at Sadore, Niger was studied during 1985 and 1986 interms of leaf area index (LAI), dry matter (DM) accumulation, net photosynthesis, stomatal conductance, total water use and yield. Among the three improved cultivars, IT82D–716 is early and erect, cv. IT83S–947 is early and spreading and cv. TVX4659–03E is a medium-duration, highyielding, dual-purpose type. The local cv. Sadore Local is a long-duration, photosensitive, spreading type used mainly for fodder. In both years, Sadore Local recorded the highest LAI. IT82D–716 and IT83S–947 produced < 1·3 t/ha of DM in both years, whereas TVX 4659–03E produced > 2 t/ha of DM and proved superior to Sadore Local in partitioning DM into pods. The four cultivars did not differ significantly either in stomatal conductance or in net phytosynthetic rates. Observed maximum photosynthetic rates of c. 20 μmol/m2/s lie at the bottom of the range 21–38 μmol/m2/s reported for 31 cowpea genotypes in an earlier study. Photosynthetic rates increased with increasing photon flux density. TVX4659–03E had an advantage in total seed plus fodder yields while the local cultivar gave significantly greater fodder yields in both years. Seed and fodder yields, as well as water-use efficiency, confirmed the advantages offered by the dual-purpose cultivar TVX4659–03E. Future breeding efforts in the Sahel should focus on dual-purpose (grain/fodder) cowpea types.

The pollination requirements of cashew (Anacardium occidentale L.), an andromonoecious tree, were studied in NE Brazil, where cashew is indigenous. It was shown through the use of bagging and caging experiments, controlled hand-pollination and emasculation of flowers, that the wind plays little role in cashew pollination and that cashew is self-fertile, although only pollen from the large stamen, either male or hermaphrodite-derived, can effect pollination. Insects appear to be the major pollinators of cashew, in agreement with the pollination syndrome suggested by the morphology of its flowers. Direct observations of insects visiting flowers and counts of pollen grains adhering to such insects suggest that the honey bee (Apis mellifera L.), although a non-native insect, is the major pollinator of commercially grown cashew in NE Brazil.

The effects of three planting patterns (ridge, furrow and flat) and three population densities (31800, 47600 and 71400 plants/ha) on the growth and yield of three sunflower hybrids, Flamme, Upsolveraflor and Isostar, were investigated at the Sultan Qaboos University Experiment Station, 1990–93. Seed and oil yields were higher in the ridge-planted sunflower than those planted in furrow or flat patterns, the order being ridge > flat > furrow. However, the magnitude of yield differences between ridge-planted and flat-planted sunflower was not sufficiently high to compensate for the additional labour and materials required for making ridges. Based on this, planting in the flat was recommended. Although increased population density resulted in taller plants, lodging was generally low and did not constitute a serious problem. Biomass/plant, head weight, receptacle weight, head diameter, number of seeds/head and 1000-seed weight decreased with increase in plant population density. Flamme had its highest seed and oil yield at 71400 plants/ha, while Isostar attained its highest yield at 47600 plants/ha. Isostar outyielded the other two hybrids and appeared to have a potential for commercial production in a desert climate.

The effect of water stress on nitrogen fixation in seven common bean (Phaseolus vulgaris L.) genotypes was investigated in Celaya, Gto., Mexico, in 1991. Beans were grown under four moisture regimes: (1) well-irrigated, control, (2) with water stress during the vegetative stage, (3) with water stress during the reproductive stage and (4) with water stress during the whole growing cycle. Biological nitrogen fixation was measured by 15N-isotope dilution using sorghum as a reference crop. Nodulation and N2-fixation data showed genotypic differences in response to water stress. Under non-stressed conditions, cv. Bayocel fixed the most nitrogen (85 kg/ha) and cultivar Flor de Mayo Baji'o the least (33 kg/ha). Under water stress at the reproductive stage, these cultivars fixed 9 and 6 kg N/ha, respectively. Water stress during the reproductive stage reduced nodulation by an average of 43% with no recovery after rewatering. Water stress during the reproductive stage had a greater effect on N2-fixation than on grain yield; in comparison to the control, N2-fixation was reduced to one sixth while grain yield was only reduced by 50%.

A grazing experiment, conducted for 22 weeks in 1992/93 at Aorangi Research Station, AgResearch Grasslands, Manawatu, New Zealand, compared the productivity of weaned lambs grazing Lotus corniculatus (birdsfoot trefoil) and lucerne (Medicago sativa). Effects of condensed tannins (CT) in lotus were evaluated by studying the responses of lambs to twice daily oral supplementation with polyethylene glycol (PEG). A rotational grazing system with restricted feed allowance was used. Measurements were made of pre- and post-grazing herbage mass, the composition of the feed on offer and diet selected, voluntary feed intake (VFI), liveweight gain (LWG), carcass growth, wool growth and the concentration of metabolites in rumen fluid. For both lotus and lucerne swards, the diet selected was mainly leaf. Lotus contained 34 g total CT/kg dry matter in the diet selected, whilst there were essentially no CT in lucerne. Compared to lambs grazing lucerne, lambs grazing lotus had slightly lower VFI, and higher LWG, carcass weight gain, carcass dressing-out percentage and wool growth. PEG supplementation had no effect on these measurements or upon the composition of rumen fluid in lambs grazing lucerne. However, in lambs grazing lotus, PEG supplementation reduced wool growth (10·9 ν. 12·1 g/day), slightly reduced LWG (188 ν. 203 g/day), increased rumen ammonia concentration, and increased the molar proportions of /.so-butyric, isovaleric and n-valeric acids and protozoa numbers in rumen fluid. PEG supplementation did not affect carcass gain, carcass fatness or the molar proportion of rumen acetic, propionic or n-butyric acids in lambs grazing lotus. It was concluded that the principal effect of CT in growing lambs grazing lotus was to increase wool growth without affecting VFI, thereby increasing the efficiency of wool production, that the greater rate of carcass gain of lambs grazing lotus than those grazing lucerne was mainly caused by factors other than CT and that CT did not affect the rumen fermentation of carbohydrate to major volatile fatty acids.

In autumn 1991, sugarbeet tops (Beta vulgaris L.) and different components of oilseed rape residues (Brassica napus L.), both labelled with 15N, were incorporated into the soil under field conditions at Askov Experimental Station, Denmark, using stainless steel cylinders to contain the treatments. The availability of this labelled N to a subsequent crop was measured, using as test crops autumn-sown rye (Secale cereale L.), wheat (Triticum aestivum L.) and spring barley (Hordeum vulgare L.). In spring 1992, cylinders with 15 N-residues received NH4NO3 and those without 15NH415NO3. In a parallel experiment, 15N-labelled beet tops were incorporated in lysimeters. A four-course rotation of sugarbeet, spring barley (undersown with perennial ryegrass Lolium perenne L.), perennial ryegrass and winter wheat at two rates of calcium ammonium nitrate (CAN) or animal slurry was grown in these lysimeters. Leaching and the availability of beet top N to successive crops were followed for 2 years. The soil in the cylinders and lysimeters was a light sandy loam (˜ 10% clay).

Of the 7·10 g N/m2 added in beet tops, 10–15% was harvested in two subsequent crops of barley and ryegrass and 13–19% was lost by nitrate leaching. Beet top N accounted for 3–7% of the total N offtake in 1992. In 1993 < 1·5% of the total N offtake in ryegrass was from the beet tops applied in 1991. Combining results from mineral fertilized treatments, it was found that 9% of the beet top N was removed in the first cereal crop, 9% was lost by nitrate leaching and 68% remained in the 0–20 cm soil layer (including roots), suggesting that the denitrification loss was < 15%.

Incorporation of oilseed rape stubble (1·35 g N/m2), two rates of pods (6·25 and 18·75 g N/m2) or mixed residues (12·25 g N/m2) contributed 0·5, 2·3, 7·4 and 4·6%, respectively, to the total N harvested in the following crop of winter wheat. The percentage of the added labelled N taken up by the wheat ranged from 4·9 to 6·1%, with 60–79% remaining in the 0–20 cm layer after harvest.

For beet tops it was calculated that 100 kg N/ha in residues incorporated in the autumn could replace 18 kg N/ha given in the following spring as mineral fertilizer. For oilseed residues, the corresponding average value was 9 kg N/ha.

In fertilized cropping systems, oilseed rape residues had minor effects on the subsequent crop, so that an uneven return of residues, as often occurs with combined crops, would do little harm. A considerable proportion of the N applied in sugarbeet tops was lost by leaching and the residual value of the sugarbeet tops to subsequent crops was low.

Among lentil-growing regions, it is only in West Asia that both large-seeded, yellow cotyledon and small-seeded, red cotyledon lentil (Lens culinaris Medikus) are grown by farmers. This study examined the adaptation to temperature and rainfall of the two seed-size groups in West Asia; first in germplasm evaluated in Syria in the 1978/79 and 1979/80 seasons and assessed for cold susceptibility in Turkey in the 1979/80 season, and second in breeding lines over the decade 1984–93 at three sites in West Asia.

Large-seeded material consistently had a longer reproductive growth period than the small-seeded group by 2·8 days, an extended period being required to fill its greater seed mass per pod. In the breeding material, the large-seeded group consistently produced taller plants and more straw than the smaller seeded-group. Germplasm with large seeds was less susceptible to winter cold than that with small seeds. Additionally, in the breeding material there was an advantage in seed yield of the largeseeded group over the small seeds in average temperatures < 10 °C from January to April, with the converse true at higher temperatures. In the breeding material, the large-seeded group showed an advantage in seed yield over their small-seeded counterparts at the two wetter sites, whereas the smallseeded group was better adapted to dry environments. Based on this adaptation, the potential of some regions currently growing exclusively large-seeded lentil for producing the other seed type is emphasised.

Six rumen-fistulated steers were studied over six periods of 3 days at Reading University's Animal Production Research Unit. In each period, eight dacron bags comprising two replicates of the 22 factorial treatment set consisting of untreated and NH3-treated barley straw in chopped (12–54 mm) or ground (2·5 mm screen) form were placed in the rumen of each steer. Bags were left in the rumen for a given time which was one of the set 6, 9, 15, 24, 48, 72 h. Percentage dry matter (DM) disappearance from dacron bags was consistently higher for NH3-treated than untreated straw and higher for ground than chopped straw. Analysis of variance gave F ratios for period and steer which were non-significant but of similar size, showing that using fewer animals could be compensated for by using more periods. Percentage DM disappearance as a function of time was well described by the model y = a — beet. The parameter defining rate of disappearance c, was similar for the four forms of straw. Differences between straws occurred in values of a (potential degradability) and b suggesting that physical form of material put into dacron bags needs standardizing if absolute values of a are to be used (e.g. for predicting roughage intake). Experimental designs need further research.

Two field experiments were conducted at Giza, Egypt in the winter seasons of 1990 and 1991, to study the effects of two fungicides (Dithane M45 and Saprol), two insecticides (Rogor and Pirimor), two acaricides (Tedifol and Comite) and two herbicides (Lancer and Cobex) on some morphological characters and yield of broad bean. The investigated pesticides were used at the recommended doses.

All pesticidal treatments reduced the percentage of missing plants at harvest. They also, except Lancer, led to an increase in the total yield of green pods and dry seeds. Individual treatments of Saprol, Dithane M45 and Cobex significantly increased the number and weight of green pods/plant and yield of dry seeds/plant. The individual treatment by Pirimor or Rogor also led to an increment in yield of dry seeds/plant. Lancer proved to be toxic to broad bean plants, as it caused a reduction in the number of branches and plant productivity of green pods and dry seeds.

The effects of dietary sucrose on the metabolic rate of plasma glucose and ruminal propionate as well as the change in nitrogen kinetics were examined in four mature wethers fitted with rumen fistulas in Tsukuba, Japan in 1990. Wethers were fed at 12 equal intervals daily on crushed lucerne hay cubes (1233 g DM/day), with or without 204 g/day of sucrose. Plasma urea and glucose kinetics were determined following a single intravenous injection of [I5N]urea and [U-13C]glucose respectively; and the kinetics of ruminal ammonia and propionate were determined following a single intraruminal injection of [15N]ammonium chloride and [2–13c]sodium propionate respectively. Following supplementation of sucrose to the diet, nitrogen retention was increased (P < 0·05) with a decrease in plasma urea concentration (P < 0·05) and urinary urea excretion (P < 0·05). Sucrose supplementation decreased (P < 005) the concentration and irreversible loss rate of ruminal ammonia. Urinary allantoin excretion did not change with sucrose treatment, but the flow rate of non-ammonia-nitrogen from the rumen was increased P < 0·05). The transfer rate of ruminal ammonia to plasma urea was also decreased (P < 0·01), whilst the transfer rate of plasma urea to ruminal ammonia was increased (P < 0·05) by dietary sucrose. Sucrose supplementation resulted in a higher concentration of propionate and butyrate (P < 0·05) in the rumen with no significant change in acetate or pH. The concentration of plasma glucose did not change with sucrose treatment, but the concentration of insulin, pool size (P < 0·05) and the irreversible loss rate of glucose (P < 0·01) were increased, reflecting the increase in the production rate of ruminal propionate (P < 0·05). It was concluded that the supplementation of sucrose affected the metabolism of urea and glucose in plasma via a change in ruminal production rate of ammonia and propionate, respectively.