包膜复合肥对夏玉米产量、氮肥利用率与土壤速效氮的影响

以郑单958为材料,比较研究了复合肥(CF)与包膜复合肥(CCF)对华北平原夏玉米产量、氮肥利用率及土壤速效氮与土壤氮素表观盈亏的影响。结果表明,1)玉米产量随施氮量增大而增大,施N.180.kg/hm2时,CCF处理高于CF处理,施N.90.kg/hm2时表现相反;2)与CF处理比较,CCF处理施N.180.kg/hm2时,氮肥利用率高5.3个百分点,施N.90.kg/hm2时低2.4个百分点;3)土壤速效氮含量一般随施氮量增大而提高,各时期表土层CCF处理较CF处理高,中、下土层表现相反。大喇叭口期之前,CF处理中、下土层土壤速效氮含量高于上土层(020.cm或040cm),而CCF180处理060.cm土层高于60120.cm土层;4)不施氮,各生育阶段均出现土壤氮素表观亏缺,且吐丝后亏缺量占总亏缺量近80%;土壤氮素表观亏缺量随施氮量增大而降低,两种肥料表现一致;同等施氮量下CCF处理亏缺量较CF处理低。包膜复合肥氮素释放较平稳,对土壤速效氮向下运移的控制较好,有利于减少氮素潜在的淋洗损失。综合考虑产量、氮肥利用与氮素损失等因素,包膜复合肥用量N.180.kg/hm2是吴桥试区夏玉米季较为理想的选择。     

不同类型氮肥与施氮量下夏玉米水、氮利用及土壤氮素表观盈亏

以两个夏播玉米品种为材料，研究了不同类型氮肥与不同施氮量下华北平原夏玉米的水分与氮素利用及土壤氮素表观盈亏．结果表明：（1）随施氮量增大．产量、耗水量及水分生产效率（WPE）增大，氮肥利用率（NUE）降低；（2）夏玉米WPE与NUE具有较明显的基因型差异。WPE郑单958较大。NUE农大108较大；（3）wPE与NUE均受到氮肥类型的影响．WPE郑单958复合肥〉普通尿素〉包膜尿素．农大108处理间差异较小，以复合肥略大，普通尿素和包膜尿素相当。与普通尿素相比，包膜尿素与复合肥利用率较高，郑单958施N90kg／hm^2与农大108施N180kg／hm^2时，两者对NUE的提高尤为明显；（4）本试验条件下，经过夏玉米生长季，两品种各处理均出现土壤氮素表观亏缺，亏缺量随施氮量增大而降低，各条件下郑单958亏缺量较农大108大；氮肥类型间土壤氮素表观亏缺量表现为复合肥〉包膜尿素〉普通尿素。     

氮肥减量后移对土壤氮素供应和夏玉米氮素吸收利用的影响

采用田间试验研究了氮肥减量后移对土壤氮素供应和夏玉米氮素吸收利用的影响。结果表明,与农民习惯施肥（N 240 kg/hm2,基肥和大喇叭口追肥为1∶2）相比,氮肥减量后移（N 168 kg/hm2,基肥、大喇叭口肥和吐丝肥为1∶3∶1）处理的产量、植株干物质积累量、植株氮积累量和积累速率均没有降低,而氮肥利用率显著增加。氮肥减量后移可使耕层无机氮供应较好地与作物吸收同步,降低收获期0—100 cm土层硝态氮积累,减少氮素的田间表观损失。基于夏玉米不同生育阶段的氮素吸收特征进行氮肥减量后移可节省氮肥30%,是较为理想的氮素施用方式。     

施氮量对冬小麦—夏玉米土壤氮素表观盈亏的影响

通过河北清苑连续6年的田间定位试验,以冬小麦—夏玉米轮作体系作为对象,设置不同的施氮处理(N_0、N_(100)、N_(180)、N_(255)、N_(330)),研究不同施氮量对冬小麦—夏玉米轮作体系的土壤氮素表观盈亏的影响。结果表明:6年土壤表观氮素累积盈亏量随施氮量的增加而增加,N_0和N_(100)处理的土壤氮素累积盈亏量为负值,N_(180)、N_(255)和N_(330)处理的土壤氮素累积盈亏量分别高达382,1 173,2 116kg/hm~2;各处理的氮素表观盈亏量年际间变异较大,而在2种作物上的氮素累积盈亏量差异不大;在冬小麦季和夏玉米季,土壤氮库达到平衡状态的施氮量分别为155,134kg/hm~2;土壤氮素表观盈亏量均与土壤无机氮变化量呈显著负相关,与施氮量呈显著正相关,且随降雨量的增加呈增加趋势;不同生育期的结果表明,在冬小麦—夏玉米生长前期均表现土壤氮素盈余,而后期土壤氮素亏缺。因此合理施用氮肥既要考虑土壤氮素盈亏,也要关注生育期的分配。     

不同类型氮肥对夏玉米产量、氮肥利用率及土壤氮素表观盈亏的影响

通过田间小区试验,比较研究了不同肥料类型、不同施氮量对夏玉米产量、氮肥利用率、土壤氮素表观盈亏量及植株氮素累积萤的影响.结果表明,夏玉米产量随施氮量增大而增大,但施肥过量产量降低,同等施氮条件下,控释掺混氮肥各处理夏玉米产量和氮肥利用率均高于普通尿素处理.施氮显著增加了0-40cm土层土壤无机氮含量,同时显著提高了玉米生长中后期植株体内氮素含量.经夏玉米季后,除高量施氮的T4,T6处理出现土壤氮索表观盈余外,其余处理均出现土壤氮素表观亏缺,施氮后土壤表观盈余量增加,且随施氮量的增加而增加.     

土壤高残留氮条件下施氮对夏玉米氮素平衡、利用及产量的影响

土壤残留氮是不容忽视的土壤氮素资源.通过田间小区试验研究了土壤高残留氮下不同施氮量(0、80、160、240和320 kg/hm2)对夏玉米土壤硝态氮积累、氮素平衡、氮素利用及产量的影响,分析了夏玉米的经济效益.结果表明,土壤剖面硝态氮积累量随施氮量的增加而增加,且施氮处理硝态氮积累量显著高于不施氮处理;各施氮处理土壤硝态氮在0-60 cm土层含量最高,在0--180 cm剖面呈先减少后增加的变化趋势.不施氮处理夏玉米收获后土壤无机氮残留量高达378 kg/hm2,随施氮量的增加,无机氮残留和氮表观损失显著增加.作物吸氮量、氮表观损失量与总氮输入量呈显著正相关,总氮输入量每增加l kg作物吸氮量增加0.156 kg,而表观损失量增加0.369 kg,是作物吸氮量的2.4倍.高残留氮土壤应严格控制氮肥用量,以免造成氮素资源的大量浪费.夏玉米籽粒吸氮量随施氮量的增加呈增加的趋势,氮收获指数呈降低的趋势.氮肥农学效率、氮肥生理利用率、氮肥利用率和氮素利用率在施氮量80 kg/hm2时最高,随施氮量的增加降低;增施氮肥能降低高残留氮土壤中氮肥的增产效果和利用率.综合考虑产量、氮素利用和环境效应,N 80 kg/hm2是氮素高残留土壤上玉米的合理施氮量.     

施氮与灌水对夏玉米土壤硝态氮积累、氮素平衡及其利用率的影响

通过田间裂区试验研究了不同施氮量(N 0、150、210和270 kg/hm2)和灌水量(900、1200、和1500 m3/hm2)对夏玉米土壤硝态氮分布累积、氮素平衡以及氮肥利用率的影响。结果表明,夏玉米收获期各处理土壤硝态氮在表层(0—20 cm)含量最高,在0—200 cm剖面均呈现先减少后增加再减少的变化趋势;土壤剖面NO3--N累积量随施氮量的增加而增加,且施氮处理硝态氮积累量显著高于不施氮处理。作物吸氮量、氮素表观损失量均与施氮量和总氮输入量呈显著相关,氮素输入量每增加1 kg,作物吸氮量仅增加0.301 kg,而表观损失量增加0.546 kg,是作物吸氮量的1.8倍左右。随施氮量的增加土壤剖面中NO3--N的损失量逐渐减少。夏玉米子粒吸氮量和收获指数随施氮量的增加有增加的趋势;氮肥回收效率和氮肥农学效率均以处理W1500N150最高,分别为46.15%和12.98kg/kg;氮肥生理效率以处理W1200N150最大,为34.49 kg/kg。本试验条件下,以水氮处理W1500N150的土壤硝态氮残留量、表观损失量较低,夏玉米氮肥回收效率和农学效率较高。     

覆膜滴灌条件下氮肥运筹对玉米氮素吸收利用和土壤无机氮含量的影响

为解决吉林省半干旱区覆膜滴灌条件下合理施氮问题,通过两年(2016—2017年)田间试验,研究了覆膜滴灌等氮量投入条件下,不同运筹模式(N1:100%基肥;N2:50%基肥+50%拔节肥;N3:30%基肥+50%拔节肥+10%大口肥+10%开花肥;N4:20%基肥+30%拔节肥+20%大口肥+20%开花肥+10%灌浆肥)对春玉米产量、氮素利用效率、关键生长节点氮素积累特征以及生育期内土壤无机氮含量变化和氮素平衡的影响。结果表明,分次施氮各处理(N2、N3、N4)玉米产量显著高于100%基肥处理(N1),其中N4处理玉米产量最高,较N1处理分别提高22.44%(2016年)和35.31%(2017年)。与N1处理相比,N2、N3、N4显著提高了玉米氮素吸收利用率、农学利用率和偏生产力,提高幅度依次为52.02%~83.21%、63.69%~120.78%、11.85%~22.46%(2016年)和92.44%~129.38%、127.23%~203.09%、22.10%~34.01%(2017年),且均以N4处理最高。施氮显著提高了玉米拔节期至成熟期氮积累量,其中开花期至成熟期氮积累量以N4处理最高。与N1处理相比,N2、N3、N4提高了玉米开花期至成熟期0~20 cm土壤无机氮含量,并降低成熟期40~100 cm土壤无机氮含量。土壤-作物系统氮素平衡中,N2、N3、N4处理较N1处理显著降低了氮素表观损失量,其中N4处理氮素表观损失量最低。综上所述,在本试验条件下,总施氮量210 kg·hm-2时,20%基肥+30%拔节肥+20%大口肥+20%开花肥+10%灌浆肥为该区域覆膜滴灌条件下氮肥最佳运筹模式。     

氮肥对吉林春玉米产量、农学效率和氮养分平衡的影响

为解决东北地区玉米合理施用氮肥问题,于2014年在吉林省中部地区通过田间试验,研究了不同施氮量(0、70、140、210、280 kg/hm2)对玉米产量、氮素吸收利用、土壤无机氮积累变化规律及氮素平衡的影响。结果表明,施氮量在70~210 kg/hm2范围内玉米产量随施氮量的增加而增加,当施氮量增加至280 kg/hm2产量下降,根据玉米产量(y)和施氮量(x)拟合得出线性加平台关系式:y=14.63x+8 734.11(R2=0.924**),得出最佳施氮量为184.0 kg/hm2。氮素利用率、农学利用率和偏生产力随施氮量的增加而下降;氮收获指数随施氮量的增加先增后降,以施氮量210 kg/hm2处理最高,为64.9%。土壤无机氮积累量在玉米整个生育期呈现先快速下降后小幅升高的趋势。玉米成熟期施氮处理各层土壤无机氮积累量均高于不施氮肥处理,且基本随施氮量的增加而增加。玉米收获后土壤无机氮残留量在施氮量70~210 kg/hm2范围内显著增加,施氮量增加至280 kg/hm2不再显著增加;氮表观损失量随施氮量的增加显著增加。玉米氮吸收量、土壤无机氮残留量和氮表观损失量与施氮量呈显著的正向相关性,玉米氮吸收量、土壤无机氮残留量和氮表观损失量分别占增加氮量的21.84%、41.19%和36.97%。综上所述,在本试验条件下,最佳施氮范围为184~210 kg/hm2。     

冬小麦-夏玉米轮作体系中不同施氮水平对玉米生长及其根际土壤氮的影响

在河北衡水潮土上进行田间试验,以当地习惯高氮用量(小麦季施N 300 kg/hm2,玉米季施N 240 kg/hm2)为对照,研究冬小麦-夏玉米轮作体系中减少氮肥用量对玉米季植株生长、氮素吸收及根际土壤中无机氮与微生物量氮的影响。结果表明,两季作物氮肥施用量减少25%和40%,对玉米产量、生物量及植株体内氮累积量未产生明显影响,氮肥利用率提高。不同氮肥施用量对根际和非根际土壤铵态氮含量的影响不显著;减少氮肥施用量,对玉米根际土壤硝态氮含量也没有明显影响。在玉米苗期、抽雄期和成熟期,习惯高施氮量处理的非根际土壤硝态氮含量较高,其中抽雄期,非根际土壤硝态氮含量较氮肥减施40%用量处理高出近一倍,但非根际土壤微生物量氮水平含量明显降低。氮肥减施未影响根际土壤微生物量碳、氮含量,反而增加了非根际土壤微生物量碳、氮水平。在高肥力的潮土上,冬小麦/夏玉米轮作体系中适当减施氮肥并未影响玉米根际土壤氮素水平,可保证玉米稳产,实现减氮增效。     

华北地区冬小麦-夏玉米轮作体系的氮素循环与平衡

对华北地区冬小麦-夏玉米轮作体系农田氮素输入输出的数量特征、平衡状况进行了分析,并评估其优化潜力。研究表明,华北地区冬小麦-夏玉米轮作体系农田每年的氮素输入中,化学氮肥、农家肥、降水、灌溉、非生物固氮和种子带入农田的氮分别为545、68、21、15、15和5kghm-2a-1,氮素年输入总量为669kghm-2a-1;每年的氮素输出中,作物收获带走的氮为311kghm-2a-1,而氨挥发、反硝化和淋洗损失的氮分别为120、16和136kghm-2a-1,氮素年输出总量为583kghm-2a-1;氮素年盈余量为86kghm-2a-1。目前我国华北地区冬小麦-夏玉米轮作体系农田氮素处于大量赢余状态,从而导致氮素大量损失。因此,加强氮肥管理,提高氮肥利用率,加大有机肥施用的力度,是华北地区农田氮素资源管理的长期任务。     

Optimum N concentration in winter wheat grown in the coastal plain region of Virginia

Abstract

Limited information is available on optimum N levels in winter wheat (Triticum aestivum L.), particularly at higher yield levels. Three experiments were conducted in the Coastal Plain region of Virginia where N was applied at rates of 0, 67, 90 and 112 kg/ha to Wheeler, Mc Nair 1003 and Coker 747 soft red winter wheat varieties. Yields ranged from 2.33 to 5.83 Mg/ha in the study. Nitrogen fertilization increased yield up to the 67 kg/ha rate and increased N concentration in the plant tissue up to 67 or 112 kg of N/ha, depending on variety. Optimum N concentration, i.e., N concentration at maximum (100%) yield for Wheeler, Mc Nair 1003 and Coker 747, over the three experiments, was 4.54%, 4.52% and 4.81%, respectively, for entire above‐ground plant samples collected at Feekes growth stage 4 and 4.72%, 4.73% and 4.44% for flag leaf samples collected at Feekes growth stage 10. A N sufficiency range of 4.00–5.00% is suggested for use for the plant parts sampled for both growth stages.     

Immobilization of ammonium and nitrate and their interaction with native N in three Illinois Mollisols

Summary Three Illinois Mollisols were incubated for 2 weeks at 25°C after treatment with different amounts of glucose and/or ¹⁵N-labelled (NH₄)₂SO₄ or ¹⁵N-labelled KNO₃. The objectives were: (1) to compare the immobilization and interaction of NH _inf4 ^sup+ –N and NO _inf3 ^sup- –N with the native soil N, and (2) to study the relationship between immobilization of applied N and the added N interaction. As determined, immobilized N refers to forms not extractable with 2 MKCl (immobilized ¹⁵N+clay-fixed ¹⁵NH _inf4 ^sup+ ). In all cases, both NH _inf4 ^sup+ –N and NO _inf3 ^sup- –N were actively immobilized and transformed into organic forms in the presence of glucose. In the absence of glucose, a higher proportion of NH _inf4 ^sup+ than NO _inf3 ^sup- was recovered in organic forms. Although the three soils differed considerably in the amounts of applied N immobilized, similar trends in N immobilization were observed. A positive added N interaction occurred with all soils, the magnitude increasing with the rate of N addition. In the absence of glucose, higher added N interactions were obtained for NH _inf4 ^sup+ than NO _inf3 ^sup- , whereas there was very little difference between NH _inf4 ^sup+ and NO _inf3 ^sup- in the presence of glucose. The results indicate that under conditions of rapid immobilization (e.g., in the presence of glucose), NH _inf4 ^sup+ and NO _inf3 ^sup- will show comparable interaction with the native soil N, whereas in unamended soil, the extent of this interaction will be greater with NH _inf4 ^sup+ than with NO _inf3 ^sup- . Significant correlations were observed between applied N immobilized and the added N interaction only in one soil having a high initial mineral N content.     

黑垆土有机氮组分对可矿化氮的关系

Mineralizable N and organic N components in different layers (0-15, 15-30, 30-45, 45-60, 60-80 and 80-100 cm) of six soils with different fertilities sampled from Yongshou County, Shaanxi Province, China,were determined by the aerobic incubation method and the Bremner procedure, respectively. Correlation,multiple regression and path analyses were performed to study the relation of minerallzable N to organic N components. Results of correlation and regression analyses showed that the amounts of the N mineralized were parallel to, and significantly correlated with, the total acid hydrolyzahle N, but was not so with the acid-insoluble N. Of the hydrolyzable N, the amino acid N and the ammonia N had a highly consistent significant correlation with the mineralized N, and their partial regression coefficients were significant in the regression equations, showing their importance in contribution to the mineralizable N. The amino sugar N, on the other hand, had a relatively high correlation with the mineralized N, but their partial regression coefficients were not significant in the regression equations. In contrast, the hydrolyzable unknown N had no such relations.Path analysis further indicated that the amino acid N and ammonia N made great direct contributions to the mineralized N, but the contributions of the amino sugar N were very low. These strongly suggested tha tthe mineralized N in the soils tested was mainly from the hydrolyzable N, particularly the amino acid N and ammonia N which are the major sources for its production.     

Mid-Season Recovery from Nitrogen Stress in Winter Wheat

ABSTRACT

Although spring-applied nitrogen (N) has been shown to be most efficient, the technique of delaying all N applications until mid-season, and the resultant effect on maximum yields, has not been thoroughly evaluated. This experiment was conducted to determine if potential yield reductions from early-season N stress can be corrected using in-season N applications. Data from three experimental sites and two growing seasons (six site-year combinations) were used to evaluate three preplant N rates (0, 45, and 90 kg ha^?1) and a range of in-season topdress N rates. Topdress N amounts were determined using a GreenSeeker hand-held sensor and an algorithm developed at Oklahoma State University. Even when early-season N stress was present (0-N preplant), N-applied topdress at Feekes 5 resulted in maximum or near-maximum yields in four of six site-year combinations when compared with other treatments receiving both preplant and topdress N.     

Comparison of Chemical Methods for Assessing Nitrogen Mineralization in Two Calcareous Soils Treated with Organic Materials

Reliable and quick methods for measuring nitrogen (N)–supplying capacities of soils (NSC) are a prerequisite for using N fertilizers. This study was conducted to develop a routine method for estimation of mineralizable N in two calcareous soils (sandy loam and clay soils) treated with municipal waste compost or sheep manure. The methods used were anaerobic biological N mineralization, mineral N released by 2 M potassium chloride (KCl), ammonium (NH₄ ⁺) N extracted by 1 N sulfuric acid (H₂SO₄), NH₄ ⁺-N extracted by acid potassium permanganate (KMnO₄), and NH₄ ⁺-N released by oxidation of soil organic matter using acidified potassium permanganate. The results showed that oxidizable N extracted by acid permanganate, a simple and rapid measure of soil N availability, was correlated with results of the anaerobic method. Oxidative 0.05 N KMnO₄ was the best method, accounting for 78.4% of variation in NSC. Also, the amount of mineralized N increased with increasing level of organic materials and was greater in clay soil than sandy loam soil.     

Apparent availability of nitrogen in composted municipal refuse

The use of composted municipal refuse on agricultural land requires prior knowledge of the interactions among compost, soil, and plants. Research into the availability of N in highly matured municipal refuse compost is particularly important considering the current concern about groundwater contamination by NO _inf3 ^sup- -N. A greenhouse pot bioassay was conducted to determine the percentage of short-term apparent bioavailable N of a highly matured refuse compost and its relative efficiency in supplying inorganic N to the soil-plant system in comparison with NH₄NO₃. Municipal refuse (after 165 days of composting) was applied at rates equivalent to 10, 20, 30, 40, and 50 t ha^-1 to a ferrallitic soil from Tenerife Island (Andeptic Paludult). NH₄NO₃ was applied at rates equivalent to the total N content of the compost treatments. Perennial ryegrass (Lolium perenne L.) was grown in 3-kg pots and the tops were harvested at regular intervals after seedling emergence. The compost increased dry matter yield, soil mineral N, and plant N uptake proportional to the applied rate. These increases were significantly higher than the control at an application rate of 20 t ha^-1. After 6 months the apparent bioavailable N ranged from 16 to 21%. The relative efficiency was 43% after 30 days. This suggests that large inputs of inorganic N into soil can be obtained with high rates of this kind of compost, with a potential for NO _inf3 ^sup- -N contamination. However, applied at moderate rates in our bioassay (<50 t ha^-1), compost showed a low N-supplying capacity to ryegrass, i.e. a small fraction of the mineralized compost N was used by plants in the course of time. This was ascribed to a partial biological immobilization. This pattern of N availability in highly matured municipal refuse compost, positive net mineralization but partial immobilization, is similar to the pattern of N availability in biologically active soils and is therefore extremely interesting for the conservation of N in agro-ecosystems.     

保护地蔬菜轮作体系中两种不同复合肥对产量的影响及其氮素动态特征

Yield and N uptake of tomato (Lycopersicum esculentum Mill.) and pepper (Capsicum annuum L.) crops in five successive rotations receiving two compound fertilizers (12-12-17 and 21-8-11 N-P₂O₅-K₂O) were studied to determine 1) crop responses, 2) dynamics of NO₃-N and NH₄-N in different soil layers, 3) N balance and 4) system-level N efficiencies. Five treatments (2 fertilizers, 2 fertilizer rates and a control), each with three replicates, were arranged in the study. The higher N fertilizer rate, 300 kg N ha^-1 (versus 150 kg N ha^-1), returned higher vegetable fruit yields and total aboveground N uptake with the largest crop responses occurring for the low-N fertilizer (12-12-17) applied at 300 kg N ha^-1 rather than with the high-N fertilizer (21-8-11). Ammonium-N in the top 90 cm of the soil profile declined during the experiment, while nitrate-N remained at a similar level throughout the experiment with the lower rate of fertilizer N. At the higher rate of N fertilizer there was a continuous NO₃-N accumulation of over 800 kg N ha^-1. About 200 kg N ha^-1 was applied with irrigation to each crop using NO₃-contaminated groundwater. In general, about 50% of the total N input was recovered from all treatments. Pepper, relative to tomato, used N more efficiently with smaller N losses, but the crops utilized less than 29% of the fertilizer N over the two and a half-year period. Local agricultural practices maintained high residual soil nutrient status. Thus, optimization of irrigation is required to minimize nitrate leaching and maximize crop N recovery.     

Organic nitrogen compounds extracted from arable and forest soils by electro-ultrafiltration and recovery rates of amino acids

Summary Extraction of synthetic amino acids dissolved in water by means of electro-ultrafiltration (EUF) showed average recovery rates of about 75%. Higher losses were obtained, particularly with cysteine, methionine and NH₄ ⁴; the latter, probably being deprotonated at the cathode, may be lost in form of NH₃. The EUF extracts of three arable and two forest soils were investigated for their N compounds. In the arable soils only about 3% of the total organic N extracted by EUF was free amino acids; about 23%–55% consisted of amino N (hydrolysable N) and the rest was non-hydrolysable N. The two forest soils contained higher amounts of EUF-extractable organic N compared with the arable soils. In the two forest soils the content of free amino-acid N amounted to 8% and 11% of the EUF organic N, and the proportion of hydrolysable N from total EUF-organic N was 41% and 46%. It is suggested that the amino-acid N and the hydrolysable N can be easily mineralized.     

Characterization of the N benefit of a grain legume (Lupinus angustifolius L.) to a cereal (Hordeum vulgare L.) by an in situ 15N isotope dilution technique

Summary A crop of barley was grown on plots which had previously supported pure stands of lupins, canola, ryegrass, and wheat. The plots were labelled with ¹⁵N-enriched fertilizers at the time of sowing of the antecedent crops. The crop of lupins, which derived 79% of its N from symbiotic N₂ fixation at physiological maturity, conferred an N benefit to barley of 3.4 g N m^-2 when compared to barley following wheat. Lupins used less fertilizer N and less unlabelled soil N compared to the other crops, but the ratios of these sources of N in the plant tops were similar. The apparent sparing of soil+fertilizer N under lupins compared with wheat was 13.6 g N m^-2, which was much larger than the measured N benefit. Barley following lupins was less enriched in ¹⁵N compared to barley following wheat, and the measured isotope dilution was used to estimate the proportion of barley N derived from biologically fixed N in the lupin residues. This in turn enabled the N benefit to be partitioned between the uptake of spared N and the uptake of fixed N derived from the mineralization of legume residues. Spared N and fixed N contributed in approximately equal proportions to the N benefit measured in barley following lupins compared to barley following wheat.