带状种植系统养分供给与利用研究进展

带状种植是1种以固N植物篱为核心、农作物与植物篱相间种植的农林复合经营模式，植物篱通过修剪避免对农作物遮光，其枝叶作为覆盖物和有机肥，可提高土壤肥力。阐述了植物篱枝叶N素矿化和利用、枝叶有机碳矿化及其对土壤有机质的影响以及固N植物篱对土壤P素和其他矿质元素的供给状况，并评价了该模式下土壤养分和有机质动态。     

Quantification of nitrogen excretion rates for three lumbricid earthworms using 15N

 Nitrogen excretion rates of ¹⁵N-labeled earthworms and contributions of ¹⁵N excretion products to organic (dissolved organic N) and inorganic (NH₄-N, NO₃-N) soil N pools were determined at 10  °C and 18  °C under laboratory conditions. Juvenile and adult Lumbricus terrestris L., pre-clitellate and adult Aporrectodea tuberculata (Eisen), and adult Lumbricus rubellus (Hoffmeister) were labeled with ¹⁵N by providing earthworms with ¹⁵N-labeled organic substrates for 5–6 weeks. The quantity of ¹⁵N excreted in unlabeled soil was measured after 48 h, and daily N excretion rates were calculated. N excretion rates ranged from 274.4 to 744 μg N g^–1 earthworm fresh weight day^–1, with a daily turnover of 0.3–0.9% of earthworm tissue N. The N excretion rates of juvenile L. terrestris were significantly lower than adult L. terrestris, and there was no difference in the N excretion rates of pre-clitellate and adult A. tuberculata. Extractable N pools, particularly NH₄-N, were greater in soils incubated with earthworms for 48 h than soils incubated without earthworms. Between 13 and 40% of excreted ¹⁵N was found in the ¹⁵N-mineral N (NH₄-N+NO₃-N) pool, and 13–23% was in the ¹⁵N-DON pool. Other fates of excreted ¹⁵N may have been incorporation in microbial biomass, chemical or physical protection in non-extractable N forms, or gaseous N losses. Earthworm excretion rates were combined with earthworm biomass measurements to estimate N flux from earthworm populations through excretion. Annual earthworm excretion was estimated at 41.5 kg N ha^–1 in an inorganically-fertilized corn agroecosystem, and was equivalent to 22% of crop N uptake. Our results suggest that the earthworms could contribute significantly to N cycling in corn agroecosystems through excretion processes. Received: 12 April 1999     

Influence of cropping intensity on the production and properties of earthworm casts in a leucaena alley cropping system

The effect of cropping intensity on surface cast production by earthworms was studied more than 4 years in an alley cropping system. The plots were cultivated to maize–cassava intercrop every year (100% cropping intensity), after 2 years of fallow (33% cropping intensity) and after 3 years of fallow (25% cropping intensity). Cast production was assessed by the continuous sampling technique proposed by Hauser and Asawalam (Z Pflanzenernahr Bodenkd 161:23–30, 1998). Results showed that cropping intensity enhanced cast production in the first year, did not affect it in the second and third years and decreased cast production in the fourth year. Amount of casts deposited within the alleys was highest under the hedgerow and decreased towards the middle of the alley. The concentrations of organic matter and plant nutrients were significantly higher in worm casts than in soil, irrespective of the cropping intensity and position of sample collection within the interrow space. Continuous cropping up to 6 years significantly reduced the concentrations of organic carbon, total nitrogen, available phosphorus, and exchangeable calcium, potassium and magnesium in the top 0–5 cm of the soil, irrespective of the distance from the hedgerow. Regardless of the intensity of cropping, there is a strong linear association between the quantity of casts and the amount of organic carbon, and other nutrients in casts.     

Uptake of 15N by wetland rice in response to application of 15N-labelled Sesbania rostrata and urea

 Pot experiments were carried out to evaluate the response of rice to Sesbania rostrata green manure N as compared to urea fertilizer N under flooded conditions. After growing S. rostrata for 21 days with a ¹⁵N-labelled N source, the labelled Sesbania was applied to wetland rice as a green manure and the uptake of ¹⁵N from this substrate was compared to that from labelled urea. Rice was cultivated twice in the same pots. The rice was grown for a period of 49 days in each case, separated by a period of 21 days when the soil was allowed to dry. The ¹⁵N content of the soil and shoots and roots of rice was determined and ¹⁵N balances established. The total N content of the shoots and roots of rice was determined by a non-tracer method. The percentage recovery of ¹⁵N from shoot material which was derived from urea N was more than twice that from S. rostrata. The recovery of ¹⁵N from the pots receiving both green manure and urea was low, and not significantly different from that recovered from the green manure treatment. As much as 64.5–73.5% and 40.1–41% of the ¹⁵N remained in the soil which had received green manure or urea, respectively. The overall recoveries of ¹⁵N varied between 86.5% and 94.4%. At the second harvest, the oven-dry weight of shoots was significantly (P<0.05) higher in green-manure treated pots, but the total N content did not differ significantly. Labelled N remaining in the soil after amendment with the green manure was much more available to the rice crop than that remaining after the addition of urea-N. The total recovery of labelled N (shoots plus roots) amounted to 65.5% and 74%, respectively of the residual labelled N in the two S. rostrata treatments (i.e. 19.55 mg ¹⁵N pot^–1 and 39.10 mg ¹⁵N pot^–1) and 23.2% and 23.2% of the residual labelled N in the two urea treatments (i.e. 19.55 mg ¹⁵N pot and 39.10 mg ¹⁵N pot^–1), respectively. Received: 8 December 1997     

Changes in 15N abundance and amounts of biologically active soil nitrogen

 Estimation of the capacity of soils to supply N for crop growth requires estimates of the complex interactions among organic and inorganic N components as a function of soil properties. Identification and measurement of active soil N forms could help to quantify estimates of N supply to crops. Isotopic dilution during incubation of soils with added ¹⁵NH₄ ⁺ compounds could identify active N components. Dilution of ¹⁵N in KCl extracts of mineral and total N, non-exchangeable NH4₄ ⁺, and N in K₂SO₄ extracts of fumigated and non-fumigated soil was measured during 7-week incubation. Samples from four soils varying in clay content from 60 to 710 g kg^–1 were used. A constant level of ¹⁵N enrichment within KCl and K₂SO₄ extracted components was found at the end of the incubation period. Total N, microbial biomass C and non-exchangeable NH₄ ⁺ contents of the soils were positively related to the clay contents. The mineralized N was positively related to the silt plus clay contents. The active soil N (ASN) contained 28–36% mineral N, 29–44% microbial biomass N, 0.3–5% non-exchangeable NH₄ ⁺ with approximately one third of the ASN unidentified. Assuming that absolute amounts of active N are related to N availability, increasing clay content was related to increased N reserve for crop production but a slower turnover. Received: 7 July 1998     

Soil N dynamics in a natural calcareous grassland under a changing climate

 This paper reports the results from a medium-term field scale investigation into the effects of simulated climate change on soil N mineralisation in a semi-natural calcareous grassland in southern England. The experiment utilised soil warming cables, automatic rainshelters and a watering system to examine two climate change scenarios: warmer winters with summer drought and warmer winters with enhanced summer rainfall. Gross N mineralisation rates in treated plots were determined, using ¹⁵N pool dilution techniques, at 6-weekly intervals over a 3-year period. Results from control plots showed a strong seasonality of mineralisation with highest rates in autumn and winter and lowest rates in summer. They suggest that water availability is the main constraint on microbial processes and plant growth. Unexpectedly, additional summer rainfall had no direct effect on N mineralisation at the time of application (summer). The treatment did, however, significantly (<0.05%) reduce rates in subsequent autumn and winter months. In contrast, summer drought significantly increased N mineralisation rates in autumn and winter. Winter warming similarly had no direct effect on N mineralisation in winter but decreased rates in spring. We hypothesise that the observed treatment effects result from changes in organic C and N input, in plant litter, resulting from the direct impact of climatic manipulation on perennial plant growth, death and senescence. This paper compares and contrasts the response to climate manipulation in the grassland system with results from other ecosystem types such as northern forests. Received: 1 December 1997     

青贮玉米-牛粪尿体系的15N标记及氮素转化研究

15N示踪技术已开始应用于畜禽粪便氮素循环与利用研究领域,而15N在畜禽粪便不同组分和不同形态氮素中的丰度与数量将直接影响到畜禽粪便15N示踪去向与氮素实际去向的一致性。为了解15N在畜禽粪便标记过程的转化特点和在标记粪尿的分布特征,本文首先采用改进的、含有15N标记硫酸铵(60 atom%15N)的Hoagland营养液砂培种植15N玉米,然后将15N玉米和普通玉米以55∶45的氮配比作为混合青贮饲料饲喂1头已空腹2 d的2龄黄牛,饲喂4 d后停喂2 d,收集全部牛粪尿并对其不同组分和形态氮素的15N丰度和数量进行分析。结果表明:标记玉米、混合青贮饲料、牛粪尿的15N丰度分别为48.024%、26.579%和8.044%;标记玉米对硫酸铵15N的回收率为26.3%,牛粪尿对标记玉米15N回收率为36.0%。在收集的牛粪尿氮中,牛粪全氮、牛尿全氮、牛粪铵态氮和牛尿铵态氮量分别占70.25%、29.75%、5.44%和0.03%,其15N丰度分别为9.223%、5.261%、6.505%和5.419%。在短期内通过饲喂黄牛15N青贮饲料制备的标记牛粪尿中,15N丰度在不同组分和形态氮素中的分布并不相同,牛尿氮的15N丰度低于牛粪氮,矿质态和易于矿化态氮的15N丰度低于不易矿化态氮。     

The use of 15N labelling to study the turnover and utilization of ruminant manure N

 An improved understanding of the cycling of animal manure N is a prerequisite for making better use of this N source. A sheep was fed ¹⁵N-labelled grass in order to study the fate of ¹⁵N-labelled ruminant manure N in the plant-soil system. The uniformity of labelling was found to be satisfactory when an appropriate feeding strategy was used. The mineralization of labelled faecal N was compared to the mineralization of labelled feed N and indigestible feed N by measuring residual labelled organic N in unplanted topsoil in the field. After 18 months, 61% of both faecal N and feed N was recovered in organic form in the topsoil, while 94% of the indigestible feed N was still present in the soil. The influence of slurry distribution in soil on the crop uptake of labelled faecal N in slurry was studied in a sandy and a sandy loam soil. The crop uptake of labelled faecal N was compared with the uptake of ¹⁵N-labelled mineral fertilizer in a reference treatment. The uptake was 28–32% of that of the reference treatment with simulated slurry injection, 13–25% with incorporated slurry and 18–19% with slurry on the soil surface. The mineralization of faecal N in the autumn after application in spring was low irrespective of the slurry distribution in soil. The results demonstrate that the contact between animal manure and the soil matrix significantly influences the short-term turnover and availability of faecal and ammonium N in slurry, especially in fine-textured soils. Received: 31 October 1997     

Spatial variation of soil enzyme activities and microbial functional diversity in temperate alley cropping systems

Spatially dependent patterns in microbial properties may exist in temperate alley cropping systems due to differences in litter quality and microclimate in areas under trees compared to those in the alleys. The effect of tree row location was evaluated for its impact on soil enzyme activities and Biolog substrate use patterns. Soils were sampled to a depth of 30 cm at the tree row and at the middle of the alley at two sites: a 21-year-old pecan (Carya illinoinensis)/bluegrass (Poa trivials) intercrop (Pecan site) and a 12-year-old silver maple (Acer saccharinum)/soybean (Glycine max)–maize (Zea mays) rotation (Maple site). Sampling was done in fall 2001 and summer 2002. β-Glucosidase activities, Biolog patterns expressed as average well color (AWC), substrate richness, and Shannon diversity index, and total Kjeldahl nitrogen (TKN) were significantly higher (P<0.05) in the tree row than at the middle of the alley for surface soils at the Pecan site. Fluorescein diacetate (FDA) hydrolytic activity was also higher at the tree row for soils sampled in the fall, but did not differ significantly for soils sampled in the summer. At the Maple site, AWC and substrate richness were significantly higher at the tree row for soils sampled in 2001. Soil volumetric water content and temperature were generally lower in the tree row at the Maple site. The results of this study suggest that functionally different microbial populations may be present under pecan trees compared to cropped alleys which may promote disparities in nutrient availability necessitating differential long-term nutrient management in such alley cropping systems.     

Influence of the presence of nitrite and nitrate in soil on maize biomass production, nitrogen immobilization and nitrogen recovery

 When comparing nitrite (NO₂ ^–) and nitrate (NO₃ ^–) toxicity to maize (Zea mays L.) growth, it is important to know the fate of applied nitrogen (N). A pot experiment, using potassium nitrite (K¹⁵NO₂) and potassium nitrate (K¹⁵NO₃) was conducted to determine the fate of N (0, 75, 150, and 225 mg N kg^–1 soil) applied to a sandy loam soil collected from Gistel (Belgium). The total dry weight of the plants treated with NO₂ ^– was lower than that of the plants treated with NO₃ ^– at 15 and 26 days after N application (harvest 1 and harvest 2, respectively). Shoot and root biomass reduction started at a relatively low NO₂ ^– application rate (75 mg NO₂ ^–-N kg^–1). Biomass reduction increased, at both harvests with increasing amounts of NO₂ ^– to more than 55% at the highest application rate (225 mg NO₃ ^–-N kg^–1). In the NO₃ ^– treatment, a reduction of 16% in total plant dry biomass was recorded only at the highest application rate (225 mg NO₂ ^–-N kg^–1), at both harvest times. The ¹⁵N plant uptake (shoots plus roots) at harvest 1 decreased with increasing N application rates of both N forms (KNO₂ and KNO₃). Twenty-six days after the N application, the total ¹⁵N taken up by the plant increased in all treatments in comparison with 15 days after the N application. However, only at higher rates of N application (150 and 225 mg N kg^–1) was the ¹⁵N uptake by the NO₂ ^– fed plants significantly lower than by the NO₃ ^– fed plants. The percentage of immobilized N from the applied N was low (0–17.7%) at both harvests, irrespective of the N source. However, with relatively low N application rates (75 mg N kg^–1), the immobilized N in the soil decreased with time. This may be due to the re-mineralization of the applied N. The percentage of inorganic ¹⁵N in the soil in NO₂ ^– treatments was slightly lower than in equivalent doses of NO₃ ^–. This might be due to higher losses of N as N-oxides. Unaccounted for N from the applied N ranged from 21% to 52% for the NO₂ ^– treatments and from 3% to 38% for the NO₃ ^– treatments. Received: 17 July 1997     

Availability of N from Casuarina residues and inorganic N to maize, using 15N isotope techniques

 The contribution of N from Casuarina equisetifolia (casuarina) residues to maize with inorganic N (ammonium sulphate) supplementation was studied in a pot experiment using ¹⁵N labelling techniques. A single rate of N application of 100 mg N kg^–1 soil was applied as N-ammonium sulphate in combination with casuarina residues in the proportions 100 : 0; 75 : 25; 50 : 50; 25 : 75 and 0 : 100, respectively. The directly ¹⁵N-labelled casuarina residue and indirect labelling (unlabelled casuarina + ¹⁵N soil) were compared to estimate the proportion and amount of N derived from the residue and fertilizer. The application of ammonium sulphate at a high rate significantly affected shoot dry matter (P<0.05) and likewise reduced the contribution of soil-derived N compared to residues. Total recoveries by maize of residue N and applied fertilizer N averaged 11% and 24%, respectively. Residue and fertilizer use efficiencies were not influenced by the addition of different rates of fertilizer or residue. The estimation of the contribution of N from different sources showed that direct measurement of the ¹⁵N-labelled organic source was more reliable. Received: 10 September 1997     

The use of 15N natural abundance variation to examine plant and soil organic fractions in pasture under different management practises

 Pasture systems lack the complexity of natural grasslands and have undisturbed soil profiles relative to arable monocultures. With controlled nitrogen (N) fertilizer inputs and measurable outputs (e.g. grazing and leaching), they can be used to investigate ¹⁵N natural abundance variation as a tool for the study of soil-N processes. In the present study, four pastures of different sward composition and N inputs were examined. Plant shoots and a range of soil fractions, categorized by size, were sampled in May prior to any major N additions, and again in July after initial N inputs had of been made. Samples were analyzed for ¹⁵N natural abundance (δ¹⁵N) and total N (εN). In the May sample plant and soil fractions varied in both ¹⁵N and εN between treatments. The 0.5 mm and 0.2 mm soil factions were comparable within treatments, as were the silt and clay fractions. Between May and July changes were apparent in the δ¹⁵N and εN of shoots and some soil fractions within each plot these corresponded to N inputs or sward type. Changes in silt-N especially, were similar to those occurring in the shoots. No comparable changes were seen in the larger fractions. Not all measured variation was explicable in this study. The inadequacies of the approach are highlighted and suggested improvements discussed. Received: 9 February 1998     

Soil microbial biomass carbon and nitrogen as affected by cropping systems

 The impacts of crop rotations and N fertilization on microbial biomass C (C_mic) and N (N_mic) were studied in soils of two long-term field experiments initiated in 1978 at the Northeast Research Center (NERC) and in 1954 at the Clarion-Webster Research Center (CWRC), both in Iowa. Surface soil samples were taken in 1996 and 1997 from plots of corn (Zea mays L.), soybeans (Glycine max (L.) Merr.), oats (Avena sativa L.), or meadow (alfalfa) (Medicago sativa L.) that had received 0 or 180 kg N ha^–1 before corn and an annual application of 20 kg P and 56 kg K ha^–1. The C_mic and N_mic values were determined by the chloroform-fumigation-extraction method and the chloroform-fumigation-incubation method, respectively. The C_mic and N_mic values were significantly affected (P<0.05) by crop rotation and plant cover at time of sampling, but not by N fertilization. In general, the highest C_mic and N_mic contents were found in the multicropping systems (4-year rotations) taken in oats or meadow plots, and the lowest values were found in continuous corn and soybean systems. On average, C_mic made up about 1.0% of the organic C (C_org), and N_mic contributed about 2.4% of the total N (N_tot) in soils at both sites and years of sampling. The C_mic values were significantly correlated with C_org contents (r≥0.41**), whereas the relationship between C_mic and N_tot was significant (r≤0.53***) only for the samples taken in 1996 at the NERC site. The C_mic : N_mic ratios were, on average, 4.3 and 6.4 in 1996, and 7.6 and 11.4 in 1997 at the NERC and CWRC sites, respectively. Crop rotation significantly (P<0.05) affected this ratio only at the NERC site, and N fertilization showed no effect at either site. In general, multicropping systems resulted in greater C_mic : C_org (1.1%) and N_mic : N_tot (2.6%) ratios than monocropping systems (0.8% and 2.1%, respectively). Received: 9 February 1999     

Long-term winter cover cropping effects on corn (Zea mays L.) production and soil nitrogen availability

 This study was conducted to determine effects of long-term winter cover cropping with hairy vetch, cereal rye and annual ryegrass on soil N availability and corn productivity. From 1987 to 1995, with the exception of the first year of the study, the cover crops were seeded each year in late September or early October after the corn harvest and incorporated into the soil in late April or early May. Corn was seeded 10 days to 2 weeks after the cover crop residues had been incorporated, and N fertilizer was applied as a side-dressing at rates of 0, 67, 134, or 201 kg N ha^–1 each year. While the average annual total N input from the above-ground biomass of the cover crops was highest for hairy vetch (72.4 kg N ha^–1), the average annual total C input was highest for cereal rye (1043 kg C ha^–1) compared with the other cover crops. Hairy vetch was the only cover crop that significantly increased pre-side-dressed NO₃ ^–-N (N_i) corn biomass and N uptake at 0 N. At an N fertilizer rate of 134 kg N ha^–1 or higher, the cover crops had a minimal effect on corn biomass. This indicated that even after 9 years of winter cover cropping, the effect of the cover crops on corn growth resulted primarily from their influence on soil N availability. The amount of available N estimated from the cover crops (N_ac) was significantly correlated with relative corn biomass production (r ²=0.707, P<0.001). The total amount of available N, comprising N_ac and N added from fertilizer (N_f), was strongly correlated (r ²=0.820, P<0.001)) with relative corn biomass production. The correlation was also high for the available N comprising N_i and N_f (r ²=0.775, P<0.001). Although cereal rye and annual ryegrass did not improve corn biomass production in the short term, they benefited soil organic N accumulation and gradually improved corn biomass production compared with the control over the long term. Received: 10 August 1999     

Leucaena hedgerow intercropping and cattle manure application in the Ethiopian highlands III. Nutrient balances

 Balances between nutrients applied or mineralized and nutrients removed in maize grain and stover were calculated in a hedgerow intercropping experiment in which Leucaena leucocephala and L. pallida prunings and cattle manure were applied. Hedgerow intercropping (also called alley cropping) is an agroforestry system in which trees are grown in dense hedges between alleys where short-cycle crops are grown. The hedges are pruned periodically during the cropping period and the prunings are added to the soil as green manure. In control treatments, nutrient depletion per season was in the order of 7–19 kg N ha^–1, 4–12 kg P ha^–1, 10–26 kg K ha^–1, 0–2 kg Ca ha^–1 and 3–6 kg Mg ha^–1. N fertilizer reversed the depletion of N, but it accelerated the depletion of the other nutrients. Manure and at least two applications of leucaena prunings resulted in net positive balances of N, K, and Ca between amounts applied or mineralized and amounts removed by maize. The amounts of P and Mg applied with, or mineralized from, prunings or manure were insufficient to offset the negative balances of these nutrients. Received: 27 January 1998     

Nitrogen and phosphorus uptake by maize as affected by particulate organic matter quality, soil characteristics, and land-use history for soils from the West African moist savanna zone

 The impact of land use (unfertilized continuous maize cropping, unfertilized and fertilized alley cropping with maize, Gliricidia sepium tree fallow, natural fallow) on the soil organic matter (SOM) status and general soil fertility characteristics were investigated for a series of soils representative for the West African moist savanna zone. Three soils from the humid forest zone were also included. In an associated pot experiment, relationships between maize N and P uptake and SOM and general soil characteristics were developed. Soils under natural fallow contained the highest amount of organic C (1.72%), total N (0.158%), and had the highest effective cation exchange capacity (ECEC) [8.9 mEq 100 g^–1 dry soil], while the Olsen P content was highest in the fertilized alley cropping plots (13.7 mg kg^–1) and lowest under natural fallow (6.3 mg kg^–1). The N concentration of the particulate organic matter (POM) was highest in the unfertilized alley cropping plots (2.4%), while the total POM N content was highest under natural fallow (370 mg N kg^–1) and lowest in continuously cropped plots (107 mg N kg^–1). After addition of all nutrients except N, a highly significant linear relationship (R ²=0.91) was observed between the total N uptake in the shoots and roots of 7-week-old maize and the POM N content for the savanna soils. POM in the humid forest soils was presumably protected from decomposition due to its higher silt and clay content. After addition of all nutrients except P, the total maize P uptake was linearly related to the Olsen P content. R ² increased from 0.56 to 0.67 in a multiple linear regression analysis including the Olsen P content and clay content (which explained 11% of the variation in P uptake). Both the SOM status and N availability were shown to be improved in land-use systems with organic matter additions, while only the addition of P fertilizer could improve P availability. Received: 9 April 1999     

Transformations of nitrogen-15-labelled urea in a flooded soil as affected by floodwater algae and green manure in a growth chamber

 The effects of floodwater algae and green manure on transformations of ¹⁵N-urea were studied in columns of a sandy loam soil in a growth chamber. The columns were flooded and either kept in the light, to allow algal growth, or in the dark (control) for 17 days before adding the labelled urea. Changes in urea-, NO₃ ^–- and NH₄ ⁺-N levels and the pH of the floodwater were measured over the subsequent 41-day period, during which the control column remained in the dark and those containing algae were maintained either in the dark to cause the death of the algae or in the light. Volatilized NH₃ was monitored, and on termination of the experiment the distribution of ¹⁵N between NO₃ ^–, NH₄ ⁺ and organic forms was measured in the soil. Urea hydrolysis was most rapid in the presence of both living algae and green manure, followed by dead algae, and was slowest in the control. The concentration of NH₄ ⁺-N in the floodwater was, however, reduced in the presence of algae due to assimilation and NH₃ volatilization owing to the raised day-time pH in the floodwater. NH₃ volatilization for the first 10 days was rather high in the columns kept in the light compared to those in the dark. Total volatilization plus denitrification losses were greatest where dead algae were present, owing to the absence of live algae which assimilated more than half of the applied N. Algal growth in floodwater increased the depth of the aerobic soil layer present at the soil-water interface. Subsequently, under dark conditions, stimulated algal growth reduced the depth of the aerobic layer causing less nitrification, which resulted in lower losses of N due to denitrification, i.e. 17% of the applied urea-N as compared to 39% in the light treatments. Although the presence of green manure caused a marked increase in the rate of hydrolysis, algal assimilation prevented excessive N losses via volatilization, indicating that the retention of higher quantities of NH₄ ⁺-N may have increased fertilizer-N use efficiency. Received: 22 January 1999     

Leucaena hedgerow intercropping and cattle manure application in the Ethiopian highlands I. Decomposition and nutrient release

 A litter bag technique was used to study the decomposition and release of N, P, K, Ca, and Mg from Leucaena leucocephala and L. pallida prunings and cattle manure in a hedgerow intercropping trial conducted in the Ethiopian highlands. Hedgerow intercropping (also called alley cropping or alley farming) is an agroforestry system in which trees are grown in dense hedges between alleys where short-cycle crops are grown. The hedges are pruned periodically during the cropping period and the prunings are added to the soil as green manure. Manure was the most resistant to decomposition, losing only 15% of its dry matter (DM) in 15 weeks, compared to 41–57% lost by leucaena prunings. Large quantities of K (up to 104 kg ha^–1) were mineralized from prunings and manure, but Ca and Mg were mostly immobilized. More N and P were released from prunings than from manure, which resulted in net immobilization of these nutrients in the initial stages of decomposition and net mineralization in later stages. Between the leucaenas more N was mineralized and less Ca and Mg were immobilized when L. leucocephala prunings were applied than when L. pallida prunings were applied. Fertilizer N increased DM decomposition and N mineralization. Mineralization of the nutrients was constrained by lignin and polyphenol contents. It is concluded that leucaena mulch and cattle manure may be significant sources of N and K for crop growth, but external sources of P, Ca and Mg may be required, particularly in acid soils which have low contents of these nutrients. However, this fertility effect has to be evaluated against the competition effect of trees to predict crop response. Received: 27 January 1997     

Natural abundance of 15N in sweet potato, pumpkin, sorghum and castor bean: possible input of N2-derived nitrogen in sweet potato

Two combinations of plant species, sweet potato (three cultivars) and pumpkin, and sweet sorghum (three cultivars) and castor bean were grown separately in three plots of alluvial soil from June to September 1996. The shoots (leaves plus stems) of sweet potato and pumpkin, and the whole tops (leaves plus stems and grains) of sweet sorghum and castor bean were harvested twice, once in August and once in September in order to analyze their natural abundance of ¹⁵N (δ¹⁵N). The δ¹⁵N values of two of the varieties of sweet potato harvested in September were significantly lower than those of pumpkin, while δ¹⁵N values of sweet potato and pumpkin harvested in August, as well as those of sweet sorghum and castor bean harvested in August and September, did not significantly differ. The lower δ¹⁵N values observed in the September-harvested sweet potato may indicate that as much as 40% of the N intake of this species is derived from dinitrogen. This species is known to have a high ability to take up N from undefined sources. Received: 23 February 1997     

Leucaena hedgerow intercropping and cattle manure application in the Ethiopian highlands II. Maize yields and nutrient uptake

 The effects of Leucaena leucocephala and L. pallida prunings and cattle manure on maize nutrient uptake and yield were investigated in a hedgerow intercropping trial in the Ethiopian highlands. Hedgerow intercropping (also called alley cropping) is an agroforestry system in which trees are grown in dense hedges between alleys where short-cycle crops are grown. The hedges are pruned periodically during the cropping period and the prunings are added to the soil as green manure. For each leucaena species, the experiment had 16 treatments resulting from a factorial combination of four levels of leucaena leaf prunings (no prunings applied; first prunings applied; first and second prunings applied; first, second and third prunings applied), two levels of air-dried cattle manure (0 and 3 t dry matter ha^–1) and two levels of N fertilizer (0 and 40 kg N ha^–1 as urea). Uptake of N, P and K increased significantly with application of the three nutrient sources, but uptake of Ca and Mg either did not respond or decreased with application of prunings and manure. All the three factors increased maize grain and stover yields significantly, usually with no significant interactions between the factors. At least two applications of prunings were required to significantly increase nutrient uptake and maize yield. Maize in the row closest to the hedge did not respond to these nutrient inputs. It is concluded that hedgerow intercropping, with or without manure application, can increase crop yields moderately (to 2–3 t ha^–1 maize grain yields) in the highlands, but P, Ca and Mg may have to be supplied from external sources if they are deficient in the soil. Additional N is still required for higher yields (>4 t ha^–1 maize grain yields). However, quantification of the competition effects of the trees is also required to confirm these results. Received: 27 January 1997