Symbiotically fixed nitrogen from field- grown white and red clover mixed with ryegrasses at low levels of15N-fertilization

A field study was carried out near Zürich (Switzerland) to determine the yield of symbiotically fixed nitrogen (¹⁵N dilution) from white clover (Trifolium repens L.) grown with perennial ryegrass (Lolium perenne L) and from red clover (Trifolium pratense L.) grown with Italian ryegrass (Lolium multiflorum Lam.). A zero N fertilizer treatment was compared to a 30 kg N/ha per cut regime (90 to 150 kg ha⁻¹ annually). The annual yield of clover N derived from symbiosis averaged 131 kg ha⁻¹ (49 to 227 kg) without N fertilization and 83 kg ha⁻¹ (21 to 173 kg) with 30 kg of fertilizer N ha⁻¹ per cut in the seeding year. Values for the first production year were 308 kg ha⁻¹ (268 to 373 kg) without N fertilization and 232 kg ha⁻¹ (165 to 305 kg) with 30 kg fertilizer N ha⁻¹ per cut. The variation between years was associated mainly with the proportion of clover in the mixtures. Apparent clover-to-grass transfer of fixed N contributed up to 52 kg N ha⁻¹ per year (17 kg N ha⁻¹ on average) to the N yield of the mixtures. Percentage N derived from symbiosis averaged 75% for white and 86% for red clover. These percentages were affected only slightly by supplemental nitrogen, but declined markedly during late summer for white clover. It is concluded that the annual yield of symbiotically fixed N from clover/grass mixtures can be very high, provided that the proportion of clover in the mixtures exceeds 50% of total dry mass yield.     

Nitrogen fixation by white clover in pastures grazed by dairy cows: Temporal variation and effects of nitrogen fertilization

Effects of rate of nitrogen (N) fertilizer and stocking rate on production and N₂ fixation by white clover (Trifolium repens L.) grown with perennial ryegrass (Lolium perenne L.) were determined over 5 years in farmlets near Hamilton, New Zealand. Three farmlets carried 3.3 dairy cows ha^–1 and received urea at 0, 200 or 400 kg N ha^–1 yr^–1 in 8–10 split applications. A fourth farmlet received 400 kg N ha^–1 yr^–1 and had 4.4 cows ha^–1.There was large variation in annual clover production and total N₂ fixation, which in the 0 N treatment ranged from 9 to 20% clover content in pasture and from 79 to 212 kg N fixed ha^–1 yr^–1. Despite this variation, total pasture production in the 0 N treatment remained at 75–85% of that in the 400 N treatments in all years, due in part to the moderating effect of carry-over of fixed N between years.Fertilizer N application decreased the average proportion of clover N derived from N₂ fixation (P_N; estimated by ¹⁵N dilution) from 77% in the 0 N treatment to 43–48% in the 400 N treatments. The corresponding average total N₂ fixation decreased from 154 kg N ha^–1 yr^–1 to 39–53 kg N ha^–1 yr^–1. This includes N₂ fixation in clover tissue below grazing height estimated at 70% of N₂ fixation in above grazing height tissue, based on associated measurements, and confirmed by field N balance calculations. Effects of N fertilizer on clover growth and N₂ fixation were greatest in spring and summer. In autumn, the 200 N treatment grew more clover than the 0 N treatment and N₂ fixation was the same. This was attributed to more severe grazing during summer in the 0 N treatment, resulting in higher surface soil temperatures and a deleterious effect on clover stolons.In the 400 N treatments, a 33% increase in cow stocking rate tended to decrease P_N from 48 to 43% due to more N cycling in excreta, but resulted in up to 2-fold more clover dry matter and N₂ fixation because lower pasture mass reduced grass competition, particularly during spring.     

The potential for breeding white clover (Trifolium repens L.) with improved nodulation and nitrogen fixation when grown with combined nitrogen

Summary Sodium nitrate applications ranging from 0.36 to 22.84 mM N were shown to depress rates of nodule formation and reduce total nitrogen fixation (acetylene reduction) in white clover plants grown in aseptic test tube culture.Low nitrate levels gave an initial depression in symbiotic activity but the reduction was of short duration and these treatments were subsequently associated with enhanced rates of nodule formation and nitrogen fixation. As a result, phenotypic variation appeared to be strongly differentially affected by the amount of nitrate present. A subsequent experiment suggested that much of the variation was a consequence of early enhancement of plant growth rates by low levels of nitrate followed by rapid depletion thus giving a transitory inhibitory effect. This was confirmed in a third experiment in which the range of nitrate concentration was held constant. Differential effects on variability in nodule formation and nitrogen fixation were then greatly reduced but there was still a residual level of plant-to-plant variation. The results have clear implications for selecting genetic variants capable of fixing di-nitrogen in the presence of combined N. The provision of a single limiting dose of combined nitrogen to a population containing individuals with inherently different growth rates can bring about variations in the phenotypic expression of symbiotic characters. These variations are unlikely to be based on genetic factors which have a direct and stable effect on nodule development and nitrogenase activity. The implications of the results for plant breeding are discussed.     

15N estimates of nitrogen fixation by white clover (Trifolium repens L.) growing in a mixture with ryegrass (Lolium perenne L.)

The ¹⁵N isotope dilution technique and the N difference method were used to estimate N₂ fixation by clover growing in a mixture with ryegrass, in a field experiment and a controlled environment experiment. Values obtained using N difference were approximately 25% lower than those estimated using ¹⁵N isotope dilution. In the field experiment there was a measured N benefit to grass growing with clover, equivalent to 42.7 kgN ha^-1. The grass in the mixture had a lower atom %¹⁵N content and a higher N content than grass in a monoculture; therefore values for N₂ fixation were different depending on choice of control plant i.e. monoculture or mixture grass. In the controlled environment experiment there were no significant differences between either the atom %¹⁵N contents or the N contents of monoculture grass and grass growing in a mixture with clover. It is concluded that there is a long term indirect transfer of N from clover to associated grass which can lead to errors in estimates of N₂ fixation.     

Interactions between nitrate uptake and N2 fixation in white clover

Summary White clover (Trifolium repens L.) plants grown in pots and supplied with the same concentration x days of¹⁵N labelled nitrate, but in contrasting patterns and doses had similar N concentrations but differed in the proportions devived from N₂ fixation and nitrate. N₂-fixation and nodule dry weight responded rapidly (2–3 days) to changes in nitrate availability. Plants exposed frequently to small doses of nitrate took up more nitrate (and hence relied less on N₂-fixation) and had greater dry weights and shoot: root ratios than those exposed to larger doses less often. In mixed ryegrass (Lolium perenne L.)/clover communities clover's ability to either successfully compete for nitrate or fix N₂ gave it consistently higher N concentrations than grass whether they were given high or low nitrate nutrient. This higher N concentration was accompanied by greater dry weights than grass in the low nitrate swards but not where high levels of nitrate were applied.     

Influence of dissimilarities in temporal and spatial N-uptake patterns on15N-based estimates of fixation and transfer of N in ryegrass-clover mixtures

The temporal N-uptake patterns of white clover (Trifolium repens L.) mixed with perennial ryegrass (Lolium perenne L.) and of red clover (Trifolium pratense L.) mixed with Italian ryegrass (Lolium multiflorum Lam.) were determined in successive harvests of herbage within the growth cycles of a ley established near Zürich (Switzerland). Rooting patterns were examined by injecting¹⁵N-fertilizer at soil depths ranging from 10 to 40 cm. The results were analyzed to determine the effect of variations in time and depth of N-uptake on the¹⁵N-based measurement of N from symbiosis (N_sym) and N from transfer (N_trans).Grasses in mixture appeared to have deeper rooting systems than grass monocultures, which led to an overestimation of N transfer from white clover to perennial ryegrass if¹⁵N was spread on the soil surface.White clover generally lagged behind grass in soil N- uptake. Soil N-uptake of red clover slowed down before that of the grass because % N_sym almost reached 100% during the second half of each growth cycle. However, the effect of these dissimilarities on the seasonal average of %N_sym did not exceed 2%.It is concluded that at the observed high levels of N₂ fixation, failure to account for the N-uptake patterns of the test and reference crops only slightly affected the estimates of % N_sym and % N_trans, and did not invalidate the observed differences between species.     

Dinitrogen fixation in white clover grown in pure stand and mixture with ryegrass estimated by the immobilized 15N isotope dilution method

Dinitrogen fixation in white clover (Trifolium repens L.) grown in pure stand and mixture with perennial ryegrass (Lolium perenne L.) was determined in the field using ¹⁵N isotope dilution and harvest of the shoots. The apparent transfer of clover N to perennial ryegrass was simultaneously assessed. The soil was labelled either by immobilizing ¹⁵N in organic matter prior to establishment of the sward or by using the conventional labelling procedure in which ¹⁵N fertilizer is added after sward establishment. Immobilization of ¹⁵N in the soil organic matter has not previously been used in studies of N₂ fixation in grass/clover pastures. However, this approach was a successful means of labelling, since the ¹⁵N enrichment only declined at a very slow rate during the experiment. After the second production year only 10–16% of the applied ¹⁵N was recovered in the harvested herbage. The two labelling methods gave, nonetheless, a similar estimate of the percentage of clover N derived from N₂ fixation. In pure stand clover, 75–94% of the N was derived from N₂ fixation and in the mixture 85–97%. The dry matter yield of the clover in mixture as percentage of total dry matter yield was relatively high and increased from 59% in the first to 65% in the second production year. The average daily N₂ fixation rate in the mixture-grown clover varied from less than 0.5 kg N ha⁻¹ day⁻¹ in autumn to more than 2.6 kg N ha⁻¹ day⁻¹ in June. For clover in pure stand the average N₂ fixation rate was greater and varied between 0.5 and 3.3 kg N ha⁻¹ day⁻¹, but with the same seasonal pattern as for clover in mixture. The amount of N fixed in the mixture was 23, 187 and 177 kg N ha⁻¹ in the seeding, first and second production year, respectively, whereas pure stand clover fixed 28, 262 and 211 kg N ha⁻¹ in the three years. The apparent transfer of clover N to grass was negligible in the seeding year, but clover N deposited in the rhizosphere or released by turnover of stolons, roots and nodules, contributed 19 and 28 kg N ha⁻¹ to the grass in the first and second production year, respectively. This revised version was published online in June 2006 with corrections to the Cover Date.     

Effect of white clover cultivar on apparent transfer of nitrogen from clover to grass and estimation of relative turnover rates of nitrogen in roots

The apparent transfer of N from clover to associated grass was evaluated over a four year period both on the basis of harvested herbage and by taking account of changes in N in stubble and root (to 10 cm depth) in swards with perennial ryegrass and three different white clover cultivars differing in leaf size. The large leaved Aran transferred 15% of its nitrogen while Huia transferred 24% and the small leaved Kent Wild White transferred 34%. When changes in stubble and root N were taken into account the percentage of N transferred was calculated to be 5% less than in harvested herbage only, as the small leaved types had proportionately more N in the roots and stolons, but the large leaved type was probably more competitive towards the grass.Loss of N from clover roots from July to October was compared to that from grass roots in a grass/white clover sward continuously stocked with steers using a method which incorporated tissue turnover and ¹⁵N dilution techniques. Less than 1 mg N m^-2 d^-1 was lost from the grass roots. In contrast 8 mg m^-2 d^-1 were estimated to be lost from clover roots while 12 mg N m^-2 d^-1 were assimilated.It is concluded that clover cultivar and competitive ability on grass have to be taken into account together with the relationship between N turnover in roots and N available for grass growth when modelling N transfer in grass/clover associations.     

Adaptation of red clover rhizobia to low temperatures

Summary Red clover Rhizobium strains, isolated from different locations between latitudes 60° and 63°30′ N in Finland, were tested for their adaptation to low temperatures. 31 strains were tested for growth at 5°C, 10°C, 15°C and 18°C in pure culture. No strain grew at 5°C. At the other temperatures there were differences between the strains, but the same strains grew fast at all temperatures. Ten strains were investigated for nodulation and acetylene reduction in phytotrons in two different climates, one simulating the growing season in southern and the other in northern Finland. There were differences between the strains in their ability to nodulate their host plant, and northern strains showed higher nitrogenase activity than southern strains in the cold climate.     

Nitrogen transfer from arrowleaf clover to ryegrass in field plantings

Arrowleaf clover (Trifolium vesiculosum Savi) and annual ryegrass Lolium multiflorum Lam.) commonly are overseeded in dormant bermudagrass (Cynodon dactylon L. Pers.) sod on coastal plain soils in the southeastern United States. Two field experiments were conducted in consecutive years at different sites to estimate the amount of N transferred from the clover to the annual grass. Nitrogen treatments included 50 kg N ha^-1 as ¹⁵N depleted ammonium nitrate applied in either February or April, and a check (no N applied). Three clippings were made during the cool-season from March to June. In both experiments, less than 5 kg N ha^-1 were transferred from the clover to the grass. Ryegrass yields of dry matter and total N were not increased by growing with clover. Clover growth was typical for the region; average dry matter yield in pure stand was 2,615 kg ha^-1 over the two-year period. Clover in mixed stand fixed between 20 and 60 kg N/ha. Less than 13% of N contained in ryegrass was transferred from arrowleaf clover to ryegrass at any clipping while clover was actively growing. The quantity of N transferred over the entire season was not statistically significant.     

Auxin-stimulated somatic embryogenesis from immature cotyledons of white clover

Cotyledons from immature embryos of white clover (Trifolium repens L.) cv. Osceola were exposed to 2,4-D or NAA to induce somatic embryogenesis. NAA at 10 or 20 mg 1^–1 was very inefficient at stimulating embryogenesis, while concentrations of 30 or 40 mg 1^–1 resulted in death of the explant tissue. Continuous exposure of cotyledons to 40 mg 1^–1 2,4-D resulted in somatic embryos which were arrested at the globular stage, or which underwent cycles of secondary embryogenesis, never proceeding beyond the globular stage. A 10 day exposure time to 2,4-D at the same concentration led to formation of somatic embryos, most of which had poorly developed cotyledons. Almost 10% of the somatic embryos converted into plants following transfer to medium devoid of growth regulators. Attempts to improve morphology of somatic embryos by using shorter exposure times to 2,4-D at 40 mg 1^–1, or by maintaining the 10 day exposure time while varying the concentration of 2,4-D, were not successful. Plants were obtained from all parents evaluated, although at different frequencies.     

Root exudates: a pathway for short-term N transfer from clover and ryegrass

The short-term transfer of nitrogen (N) from legumes to grasses was investigated in two laboratory studies. One study was done in pots where the roots of white clover (Trifolium repens L.) and perennial ryegrass (Lolium perenne L.) were allowed to co-exist, and a second study was performed using a micro-lysimeter system designed to maintain nutrient flow from the clover to the grass, whilst removing direct contact between the root systems. The ¹⁵N-dilution technique was used to quantify the transfer of N between species. Levels of ammonia and amino acids were measured in root exudates. The amounts of N transferred were in the same order of magnitude in both the pot and micro-lysimeter experiments. In the micro-lysimeter experiment, 0.076 mg of N were transferred per plant from clover to ryegrass during the course of the experiment. Ammonium exudation was much higher than amino acid exudation. The most abundant amino acids in both clover and ryegrass root exudates were serine and glycine. However, there was no correlation between the free amino acid profile of root extracts and exudates for both plant species: Asparagine was the major amino acid in clover roots, while glutamine, glutamate and aspartate were the major amino acids in ryegrass roots. Comparison of exudates obtained from plants grown in non-sterile or axenic conditions provides evidence of plant origin of ammonium, serine and glycine.     

Variation for aluminium tolerance in white clover

Aluminium (Al) tolerance of fourteen white clover (Trifolium repens L.) cultivars from eleven countries was compared in the greenhouse in the Wainui silt loam (Typic Dystrochrept) to which Al had been added at nine levels (0, 2.5, 5, 20, 50, 150, 250, 500 and 750 mg kg⁻¹ of soil) as Al₂ (SO₄)₃ and incubated for 30 days. None of the white clover cultivars, including those either referred to as Al-tolerant, Dusi and Pathfinder, or from countries that have large areas of acid soils, El Lucero M.A.G., Bayucua, Bage and Zapican, showed greater Al-tolerance than ‘Grasslands Huia’ white clover. Subsequent screening for Al-tolerance can therefore be restricted to germplasm with wide agronomic adaptation.     

The response of white clover (Trifolium repens L.) seedlings to spring root temperatures: The relative roles of the plant and the Rhizobium bacteria

Three experiments are reported which examine the relative roles of host and Rhizobium genotypes as factors limiting clover (Trifolium repens L.) growth at low soil temperatures.In the first experiment un-nodulated clover and perennial ryegrass (Lolium perenne L.) were grown with non-limiting nitrate at root temperatures of 8, 10 and 12°C. The ryegrass had substantially better relative growth rates (RGR) than the clover with the biggest difference occurring at 8°C. Alterations in growth rate with temperature were more marked in clover than in ryegrass but the latter still produced several times more dry matter than clover at each temperature.In the subsequent experiments clover nodulated with different strains of rhizobia was grown with and without non-limiting additions of nitrate at root temperatures of 9, 12 and 15°C. Plants receiving nitrate generally produced more dry matter than those dependent upon Rhizobium for nitrogen but differences in yield between these treatments did not alter with temperature. This suggests that limitations imposed by nitrogen fixation are similar at both high and low temperatures. Indeed, there was some evidence that nitrogen limitations were rather more pronounced at the highest temperature. The first experiment clearly demonstrated that the clover genotype makes particularly poor use of nitrate at low root temperatures when compared to its common companion perennial ryegrass.It can be concluded that improvements in spring growth of clover will rest largely with alterations to the plant genotype and its ability to use combined nitrogen for growth at lower temperatures rather than with changes in rhizobia or any symbiotic characters.     

The growth and asymmetry of neighbouring plants of white clover (Trifolium repens L.)

Summary Transplants of white clover (Trifolium repens L.) were grown isolated from each other and in pairs placed at different distances apart. The paired plants developed asymmetrically and at the interface between paired clones both the density of nodes and of stolons appeared to reach ceiling values that were of the same order as those achieved in isolated clones. It is argued that the growth of plants of T. repens is controlled by the local conditions experienced by the plant parts and not by integrated growth of the whole. Transplants of three different genotypes of T. repens, which differed in growth form, were grown as neighbouring pairs and the calculated asymmetry of the plants was used to compare their mutual aggressivenes. The more compact (phalangeal) genotypes induced greater asymmetry in their neighbours than the more diffuse forms.     

Plant factors influencing phosphorus uptake by white clover from solution culture

Summary The phosphorus (P) uptake rate of several white clover populations was determined in two solution culture experiments. Populations and cultivars differed in P uptake per plant and per unit root length in both experiments. Correlation and multiple regression analysis showed that differences between populations for P uptake per plant were largely related (r²>80%) to differences in leaf area and absolute growth rate, when plants had been grown at high-P levels, and by differences in root size and absolute growth rate when plants had been grown at low-P levels. Differences between populations for P uptake per unit root length were related (r²≈50%) to leaf area and relative growth rate in experiment 1 and to transpiration rate and water influx in experiment 2, when plants were pretreated at high-P levels. Differences between populations for P uptake per unit root length were negatively related to root size when plants had been grown at low-P levels. On the basis of these and other results it is suggested that P uptake per plant is determined largely by shoot factors. However, P uptake per unit root length is negatively related to root size, because demand for P is largely determined by shoot factors, and so differences in root size lead to an apparent difference in uptake per unit of root size.     

Nitrogen fertilization effects on dinitrogen fixation as influenced by legume species and proportion in legume-grass mixtures in Uruguay

Grasses grown in mixture with nodulated legumes often are N-limited, but N fertilization may result in reductions of N₂ fixation and legume stands. We studied N-fertilizer effects on N₂ fixation for three binary legume-grass mixtures in Uruguay. Replicated swards of white clover (Trifolium repens L.) (WC), red clover (Trifolium pratense L.) (RC), or birdsfoot trefoil (Lotus corniculatus L.) (BT), each in combination with tall fescue (Festuca arundinacea Schreb) (TF) at two legume proportions were sown in 1983 (Exp. 1) and 1984 (Exp. 2). In the fall of 1984, N treatments at 100 kg ha⁻¹ and controls were randomly assigned to subplots in Exp. 1 (established swards) and in Exp. 2 (at seeding). The soil for both experiments was a fine, montmorillonitic, mesic, Typic Argiudolls. Herbage fixed-N was estimated by ¹⁵N isotope-dilution with pure stands of TF as reference. In both experiments, N fertilization reduced the proportion of legume N derived from air (% Ndfa) and increased herbage yield only during the first 18 to 20 weeks after application. Fertilizer-N reduced annual fixed-N yield from 178 to 148 kg ha⁻¹ in Exp. 1 and from 65 to 29 kg ha⁻¹ in Exp. 2 Fixed-N yield for BT was markedly reduced by N in both experiments (33 to 53%), whereas for the clovers reduction was lesser in Exp. 1 (9 to 13%) than in Exp. 2 (46 to 64%). Negative effects of N on % Ndfa were more evident for the high legume proportion. We conclude that fertilization with 100 kg N ha⁻¹ reduced % Ndfa only for the immediate 18 to 20 weeks after application. Fertilizer-induced reduction of fixed-N yields lasted longer because of a more prolonged depression of legume proportion, especially for BT and for newly seeded swards. Journal Paper no. J.-13327 of the Iowa Agric. and Home Econ. Exp. Stn., Ames, U.S.A. (Project 2281). Supported in part by the Facultad de Agronomía, Montevideo, Uruguay; and the International Atomic Energy Agency, Vienna, Austria (Project URU/5/012). Journal Paper no. J.-13327 of the Iowa Agric. and Home Econ. Exp. Stn., Ames, U.S.A. (Project 2281). Supported in part by the Facultad de Agronomía, Montevideo, Uruguay; and the International Atomic Energy Agency, Vienna, Austria (Project URU/5/012).     

Long- and short-term effects of crop residues on aluminum toxicity,phosphorus availability and growth of pearl millet in an acid sandy soil

Pasture swards containing perennial ryegrass (Lolium perenne L.) alone or with one of five different white clover (Trifolium repens L.) cultivars were examined for production and transfer of fixed nitrogen (N) to grass under dairy cow grazing. Grass-only swards produced 21% less than mixed clover-grass swards during the second year after sowing. Production from grass-only plots under a mowing and clipping removal regime was 44% less than from grass-only plots under grazing. Much of this difference could be attributed to N transfer. In swards without clover, the ryegrass component also decreased in favour of other grasses.The average amount of fixed N in herbage from all clover cultivars was 269 kg N ha^–1 yr^–1. Above-ground transfer of fixed N to grasses (via cow excreta) was estimated at 60 kg N ha^–1 yr^–1. Below-ground transfer of fixed N to grasses was estimated at 70 kg N ha^–1 yr^–1 by ¹⁵N dilution and was similar for all clover cultivars. Thus, about 50% of grass N was met by transfer of fixed N from white clover during the measurement year. Short-term measurements using a ¹⁵N foliar-labelling method indicated that below-ground N transfer was largest during dry summer conditions.     

Biological nitrogen fixation in grass–clover affected by animal excreta

Aiming at estimating the average N2-fixation in a pasture, ap preciating the great variability due to patchy urine and dung deposition, the in fluence of dairy cow excreta on biological N2-fixation in a perennial ryegrass–white clover mixture was studied using natural urine and dung. Application of urine as well as dung affected the N2-fixation by promoting the growth of grass and thereby the proportion of clover was significantly reduced. Also the proportion of clover-N derived from the atmosphere (pNdfa) was significantly reduced. In control plots clover dry matter constituted between 40 and 50% of the total dry matter production and the pNdfa ranged between 0.8 and 0.9. Addition of urine caused a significant increase in the grass growth rates, which was the primary reason for a decrease in proportion of clover. At the same time pNdfa decreased to 0.2–0.4 followed by an increase resulting in a total reduction of 45% in the N2-fixation in urine affected areas over a period of four months. The dung only affected the N2-fixation for a distance of up to 10 cm from the edge of the dung pats. In this border area the pNdfa decreased from 0.85 to 0.75 during one month after application followed by an increase, so that after three months there was no difference between pNdfa at 0–10 and 10–20 cm distance from the dung hill. The proportion of clover was lower in the 0–10 cm than in the 10–20 cm distance, which totally resulted in a total reduction of 20% in the N2-fixation over a period of four months in the 0–10 cm area around the dung pats. Considering the proportion of a pasture which may by affected by excreta at a stocking density of 4–6 cows ha-1, the length of the grazing period, the frequency of excretion and the area covered by individual patches, it was estimated that the N2-fixation in a grass-clover pasture would be reduced by 10–15% compared to the N2-fixation in a grass-clover sward not exposed to animal excreta.     

Nitrogen fixation in perennial forage legumes in the field

Nitrogen acquisition is one of the most important factors for plant production, and N contribution from biological N₂ fixation can reduce the need for industrial N fertilizers. Perennial forages are widespread in temperate and boreal areas, where much of the agriculture is based on livestock production. Due to the symbiosis with N₂-fixing rhizobia, perennial forage legumes have great potential to increase sustainability in such grassland farming systems. The present work is a summary of a large number of studies investigating N₂ fixation in three perennial forage legumes primarily relating to ungrazed northern temperate/boreal areas. Reported rates of N₂ fixation in above-ground plant tissues were in the range of up to 373 kg N ha^–1 year^–1 in red clover (Trifolium pratense L.), 545 kg N ha^–1 year^–1 in white clover (T. repens L.) and 350 kg N ha^–1 year^–1 in alfalfa (Medicago sativa L.). When grown in mixtures with grasses, these species took a large fraction of their nitrogen from N₂ fixation (average around 80%), regardless of management, dry matter yield and location. There was a large variation in N₂ fixation data and part of this variation was ascribed to differences in plant production between years. Studies with experiments at more than one site showed that also geographic location was an important source of variation. On the other hand, when all data were plotted against latitude, there was no simple correlation. Climatic conditions seem therefore to give as high N₂ fixation per ha and year in northern areas (around 60°N) as in areas with a milder climate (around 40°N). Analyzing whole plants or just above-ground plant parts influenced the estimate of N₂ fixation, and most reported values were underestimated since roots were not included. Despite large differences in environmental conditions, such as N fertilization and geographic location, N₂ fixation (Nfix; kg N per ha and year) was significantly (P<0.001) correlated to legume dry matter yield (DM; kg per ha and year). Very rough, but nevertheless valuable estimations of Nfix in legume/grass mixtures (roots not considered) are given by Nfix = 0.026DM + 7 for T. pratense, Nfix = 0.031DM + 24 for T. repens, and Nfix = 0.021DM + 17 for M. sativa.