Residual effect of long-term applications of farmyard manure to silage maize

Sluijsmans and Kolenbrander developed a simple model to describe the availability of animal manure, assuming a readily available, an easily decomposable and a slowly decomposable N fraction. We tested this model on data from an experiment in which farmyard manure had been applied for eleven successive years to silage maize [Zea mays L.] grown on a light sandy soil. The residual effects of this FYM were then measured by growing Italian ryegrass [Lolium multiflorum Lamk.] in the 12th year. The measured uptake of N by the grass of the FYM residues was then compared with the computed values. The measured amounts of N taken up agreed fairly well with the calculated amounts for applications of 50 and 100 t FYM per ha per year.If the rates of manure application are adjusted to crop requirement, the model shows that the potential, long-term release of N from the residual N fraction of FYM will not exceed 20 kg N per ha. For cattle slurry with a smaller residual fraction, the release will be at most 10% of the total annual N application.     

Urea management for lowland transplanted rice as affected by application of farmyard manure

Farmyard manure (FYM) applied to rice-growing soils can substitute for industrial fertilizers, but little is known about the influence of FYM on the effectiveness and optimal management for industrial N fertilizers. A field experiment was conducted in northern Vietnam on a degraded soil in the spring season (February to June) and summer season (July to November) to determine the effect of FYM on optimal timing for the first application of urea. The experimental design was a randomized complete block with two rates of basal incorporated FYM (0 or 6 Mg ha^–1) in factorial combination with two timings of the first application of 30 kg urea-N ha^–1 (basal incorporated before transplanting or delayed until 14 to 16 d after transplanting). The FYM was formed by composting pig manure with rice straw for 3 months. Basal incorporation of FYM, containing 23 kg N ha^–1, increased rice grain yield in both seasons. The yield increase cannot be attributed to reduced ammonia loss of applied urea-N, because FYM did not reduce partial pressure of ammonia (pNH₃) following urea application in either season. Basal and delayed applications of urea were equally effective in the absence of FYM, but when FYM was applied rice yields in both seasons were higher for delayed (mean = 3.2 Mg ha^–1) than basal (mean = 2.9 Mg ha^–1) application of urea. Results suggest that recommendations for urea timing in irrigated lowland rice should consider whether farmers apply FYM.     

Emissions of NH3, N2O and CH4 from dairy cows housed in a farmyard manure tying stall (housing, manure storage, manure spreading)

Emission measurements from dairy cows housed in a tying stall were carried out with the aim of finding factors that influence the amount of emissions and means to reduce emissions. All sectors of animal husbandry were investigated. This enabled calculations of emissions for the whole management system including housing, storage and spreading of manure. Emissions during aerobic composting and anaerobic stacking of farmyard manure were compared. NH₃ and N₂O emissions from tying stalls for dairy cows are low (5.8 g NH₃ LU⁻¹ d⁻¹, 619.2 mg N₂O LU⁻¹ d⁻¹). Methane emissions from the animal housing are mainly caused by enteric fermentation. During storage and after spreading of farmyard manure substantial differences concerning NH₃, N₂O and CH₄ emissions were observed with composted and anaerobically stacked farmyard manure. The compost emitted more NH₃ than the anaerobically stacked farmyard manure. About one third of the NH₃ emissions from the anaerobically stacked farmyard manure occurred after spreading. Total N losses were at a low level with both storage systems. Greenhouse gas emissions (N₂O and CH₄) were much higher from the anaerobically stacked farmyard manure than from the composted one. As these are ecologically harmful gases, they have to be considered when judging the form of manure treatment. This revised version was published online in August 2006 with corrections to the Cover Date.     

Phosphorus management of a rice-wheat cropping system

A long term field experiment was conducted on a sandy loam soil from 1983 to 1987 to determine how to best apply phosphorus fertilizer in a rice-wheat cropping system. The treatments included 9 combinations of phosphorus application either to both rice and wheat or to rice or wheat alone. Direct application of phosphorus at 13 kg/ha to both the crops resulted in significantly higher total productivity of the rice-wheat cropping system as compared with 26 kg P/ha applied either to rice or wheat alone. Phosphorus at 13 kg/ha for rice and 26 kg/ha for wheat was as efficient as 13 kg P/ha for rice and 13 kg P/ha for wheat. The higher rate of P (26 kg/ha) applied to both rice and wheat resulted a decline in the total productivity. The residual effects of phosphorus applied to either rice or wheat were significant to the succeeding crop but was inferior to its direct application. Phosphorus increased the leaf area index, chlorophyll content of leaves, and interception of more photosynthetically active radiation (PAR) which resulted in increased grain yield of rice and wheat. Phosphorus status of the surface soil declined markedly, in the absence of P application from 15.4 to 6.4 kg P/ha. Phosphorus applied at 26 kg P/ha to both the crops resulted a build up of the available P status of soil. Phosphorus application at 13 kg/ha to both rice and wheat maintained the phosphorus status of the soil at original level.     

The value of poultry manure for wetland rice grown in rotation with wheat

Poultry manure applied alone or in combination with urea at different N levels was evaluated as a N source for wetland rice grown in a Fatehpur loamy sand soil. Residual effects were studied on wheat which followed rice every year during the three cropping cycles. In the first year, poultry manure did not perform better than urea but by the third year, when applied in quantities sufficient to supply 120 and 180 kg N ha^–1, it produced significantly more rice grain yield than the same rates of N as urea. Poultry manure sustained the grain yield of rice during the three years while the yield decreased with urea. Apparent N recovery by rice decreased from 45 to 28% during 1987 to 1989 in the case of urea, but it remained almost the same (35, 33 and 37%) for poultry manure. Thus, urea N values of poultry manure calculated from yield or N uptake data following two different approaches averaged 80, 112 and 127% in 1987, 1988 and 1989, respectively. Poultry manure and urea applied in 1:1 ratio on N basis produced yields in between the yields from the two sources applied alone. After three cycles of rice-wheat rotation, the organic matter in the soil increased with the amount of manure applied to a plot. Olsen available P increased in soils amended with poultry manure. A residual effect of poultry manure applied to rice to supply 120 or 180 kg N ha^–1 was observed in the wheat which followed rice and it was equivalent to 40 kg N ha^–1 plus some P applied directly to wheat.     

Effects of organic manure on solubility and mobility of different phosphate fertilizers in two paddy soils

The solubility and mobility of three phosphate fertilizers [superphosphate (SP), Ca-Mg phosphate (Ca-Mg-P) and diammonium phosphate (DAP)] with or without combination of farmyard manure (FYM) and/or farmyard manure juice (FYMJ) in two paddy soils (Ultisol and Entisol) from the Zhejiang Province of China were determined. Incubation experiments were conducted under controlled greenhouse conditions in the Institute of Agricultural Chemistry, Bonn University, Germany. The results showed that water soluble P and Olsen P in both the application band and the adjacent zones (10 and 20 mm from the application band) was highest with DAP, and lowest with Ca-Mg-P. Both FYM and FYMJ increased water soluble P (100–300%) and Olsen P (80–300%) for the application band in both soils. Moreover, the combination of FYM significantly increased water soluble P and Olsen P for the adjacent zones (20 mm for the Ultisol and 10 mm for the Entisol), but such effects were not significant in combination with FYMJ. Consequently, the main effect of FYM is due to organic substances from decomposition rather than organic compounds in the FYMJ. Also, the combination of FYMJ with SP and DAP significantly increased the pH value in the application band. The pH in the adjacent zone (10 mm) was slightly increased by the combination of FYM with each of three P fertilizers. The solubility and mobility of phosphate in soils differed greatly among phosphate fertilizers. The combinations of organic manure with mineral phosphate fertilizers increased solubility and mobility of phosphate in both paddy soils, especially in the Ultisol.This paper is dedicated to Professor Xi Sun.     

Short-term carbon dioxide emissions and denitrification losses from soils amended with low-P manure from genetically modified pigs

Genetically modified pigs have been developed that are able to use phosphorus (P) in the feed more efficiently and thereby reduce the amount of P excreted in the manure. Improved digestibility would also change the chemical composition of the manure. The objective of this study was to determine how improved P digestibility affects the manure composition and the associated carbon dioxide (CO₂) emissions and nitrogen (N) loss through denitrification when the manure is applied to soil. Aerobic and anaerobic incubation studies were conducted to evaluate CO₂ emissions and denitrification losses from two soils (Brookston clay loam and Harrow sandy loam) amended with pig slurry (115 mg N kg⁻¹ soil) from transgenic pigs as well as from conventional pigs. In addition, both the transgenic and conventional pigs were fed either a low-P diet or a conventional P diet, and the effects of diet type on CO₂ emissions and denitrification losses were examined. Carbon dioxide emissions were 17% lower (P < 0.05) in treatments amended with transgenic pig manure compared with conventional pig manure in the clay loam soil. However, denitrification losses were increased by 37% (P < 0.05) in the clay loam soil amended with manure from transgenic pigs compared to manure from conventional pigs. Neither CO₂ emissions nor N loss through denitrification were affected by pig genotype when the manure was added to the sandy loam soil. The diet type (conventional P diet or low-P diet) did not affect either CO₂ emissions or N losses through denitrification in either the Brookston clay loam or Harrow sandy loam soils.     

Nitrate leaching from loamy soils as affected by crop rotation and nitrogen fertilizer application

Nitrate leaching as affected by cropping system/crop rotation, history of farmyard manure application or fertilizer nitrogen application (0 N, 0.5 N and 1 N) was studied at nine sites on loamy soils during 1986/87, 1987/88 and 1988/89. Soil solution from 80 to 90 cm soil depth was sampled every second week in the period November to May by the use of porous ceramic cups and analysed for NO₃-N and Cl. Climatical conditions were obtained from standard meteorological observations in the region. Drainage from soil profiles was calculated from measured and simulated values of precipitation and actual evapotranspiration, respectively.The results show that type of crop is of the utmost importance for the leaching magnitude of nitrate as 40% of the total variance in nitrate concentrations in the soil solution could be explained by the type of crop.The second factor of importance was the history of farmyard manure (FYM) application, which was able to explain 28% of the total variation in nitrate concentration in the soil solution. Nitrate concentration/leaching from arable land without FYM ever being applied was considerably lower than from arable land which received periodical FYM applications until the early 70's or from arable land which besides periodical FYM applications in the past presently still receives regular applications of FYM. Only about 1% of the total variation in nitrate concentration in the soil solution was explainable by the level of fertilizer nitrogen application.Differences between years explained 14% of the total variation in nitrate concentration in the soil solution due to differences between the years in soil temperatures and water run-off. The run-off during the autumn and winter of 1986/87, 1987/88 and 1988/89 was 95, 275 and 55 mm, respectively. As expected nitrate leaching increased with increasing run off. However, nitrate leaching at the 275 mm run off was considerably lower than expected, which seems explainable by a substantial denitrification. The soil at the sites in question seems thus partly to purify the soil solution for nitrate before it leaves the root zone at the observed high run off conditions.     

Phosphorus availability to irrigated lowland rice as affected by sources,application level and green manure

To evaluate alternative fertilizer phosphorus (P) sources in lowland rice, two field experiments were conducted under irrigated conditions in Quezon Province, Philippines during 1990–1991 crop year. In another field experiment fertilizer P recycling through a green manure crop applied in the succeeding rice, was studied. Addition of fertilizer P increased grain yield by 1.5–2.0 t/ha (46%) in 1990 wet season (WS) and by 1.6–2.1 t/ha (56%) in 1991 dry season (DS). However, fertilizer P source and application level did not effect grain yield significantly. Results indicated that the less water-soluble and less expensive partially acidulated phosphate rock (PAPR), phosphate rock (PR) and less reactive PR were as effective as the more soluble but more expensive triple superphosphate (TSP). The relative effectiveness (RE) of local guano was significantly lower than that of other sources of fertilizer P. Fertilizer P applied to a pre-rice Sesbania rostrata green manure increased rice grain yield by 1.5–1.9 t/ha during 1991 DS. Further, S. rostrata fertilized with Morocco phosphate rock (MPR) gave significantly higher rice grain yield than did rice fertilized with MPR applied alone. In the P source experiments Olsen method and Pi correlated better with growth attributes than Bray 2 P. Phoshours uptake did not differ significantly among P sources and levels. Results suggest that P uptake was improved with green manuring. Correlation analyses revealed a close correlation between P uptake and dry matter yield and P uptake and grain yield.     

The phosphorus requirement of different crop species compared with wheat on lateritic soils

The phosphorus (P) requirement for grain production of different crop species (oats (Avena sativa), barley (Hordeum vulgare), triticale (xTriticosecale), narrow-leafed lupins (Lupinus angustifolius), and sandplain lupins (L. cosentinii) was compared with wheat (Triticum aestivum) in five field experiments on different lateritic soils in south-western Australia. Seven or eight levels of superphosphate were applied at the start of each experiment. The amount of P required to produce 70% (four experiments) or 90% (one experiment) of the maximum yield was used to compare P requirements. Large differences in the P requirements of the species were obtained.On P deficient soil in 3 experiments, oats required from 50 to 70% less P than wheat, but required 40% more P on a soil with a long history of superphosphate applications.Compared with wheat, in the year of P application, barley required 50% less P in one experiment, had similar P requirements in two experiments, and required 80% more P in another experiment. In the years after P application, barley required 20% less P in two experiments.On an acidic soil triticale required from 50% to 70% less P than wheat, but on less acidic soil it required 100% more P.In the year of P application, narrow-leafed lupins required 800% more P than wheat in one experiment, and 30% more P in the other experiment.In the year of P application, sandplain lupins required 70% less P than wheat in one experiment.     

Inorganic phosphorus and manure effects on bahiagrass production on a Spodosol

Bahiagrass (Paspalum notatum Flugge) pasture fertilization recommendations have traditionally been based upon clipping studies. Inclusion of P from manure, not originally considered when P recommendations were developed for pastures, may minimize the need for P fertilization without reducing bahiagrass production or P uptake. The objective of this research was to determine if manure contributes greatly to the P crop nutrient requirement. A 2-year field study utilized a factorial arrangement of 0 and 6.9 Mg air-dried manure ha^-1 with 0, 17, 34, 51, and 68 kg inorganic P ha^-1 from triple superphosphate to evaluate bahiagrass yield, root distribution, and P uptake response on a Myakka fine sand (sandy, siliceous, hyperthermic Aeric Alaquod). Because air-dried manure was used in the field study, a greenhouse study was employed to confirm that there were no differences in bahiagrass yield or P uptake from either air-dried or fresh cattle (Bos spp.) manure sources. There were no manure or manure by P interaction effects on yield or P uptake of bahiagrass indicating that manure source did not effect grass production in the greenhouse. In the field study, bahiagrass roots were distributed into the Bh horizon, and the Bh horizon had at least four times more Mehlich-1 extractable P than that of the Ap horizon. This horizon was most likely acting as a main source for P-uptake by the grass. This observation was further confirmed by no yield response to levels of inorganic P application in 1989. A linear-response-and-plateau (R²=0.196) relationship with a critical point of 15.4 kg P ha^-1 was found in 1990. Bahiagrass yield and P uptake were not dependent on P fertilization, either from manure or inorganic P, due to the availability of P from the Bh horizon.     

An economic evaluation of a long-term experiment on phosphorus and manure amendments to sandy Sahelian soils: Using a stochastic dominance model

Poor fertility status of sandy Sahelian soils represents a major constraint to cereal and legume production. Soil amendment options were evaluated, using a stochastic efficiency framework. Dominance analyses showed that in the presence of annual applications of 30 kg N ha^–1 and 30 kg K ha^–1, efficient soil amendment options comprise of either the annual application of 8.7 kg P ha^–1 in the form of single superphosphates in combination with 5 tonnes manure ha^–1 applied every three years or the annual application of 17.5 kg P ha^–1 in the form of single superphosphates. Choice between these two efficient options depends on the availability of manure, deficiencies in sandy soils and farmer resource endowments.Submitted as JA no. 1133 by the International Crops Research Institute for the Semi-Arid Tropics (ICRISAT).     

Efficiency of N-(n-butyl) thiophosphoric triamide in retarding hydrolysis of urea and ammonia volatilization losses in a flooded sandy loam soil amended with organic materials

Using a forced-draft chamber technique, the suppression of NH3 volatilization losses by applying N-(n-butyl) thiophosphoric triamide (NBPT) was studied in an alkaline sandy loam soil amended with green manure or wheat straw. Applied urea was completely hydrolysed in 12, 8 and 6 days in unamended, green manure and wheat straw amended soil, respectively. By applying 0.5% (w/w of urea) NBPT, complete hydrolysis of urea was delayed up to 16 days in the unamended soil, whereas in wheat straw amended soil urea hydrolysis was completed by the 12th day even when it was treated with 2% NBPT. Applied at 1 or 2% level, NBPT delayed the NH3 volatilization to the 4th day after application of urea in green manure or wheat straw amended soil. Hydrolysis of urea was more effectively retarded by applying NBPT in the unamended soil than in soil amended with green manure or wheat straw. In the unamended soil, 7.1% of the applied urea was lost through NH3 volatilization. The losses were reduced to 1.2 and 0.7% by applying 0.5 and 1% NBPT, respectively. Enhanced NH3 volatilization caused by the green manure or wheat straw was counteracted by applying NBPT.     

Transformations of soil and manure phosphorus after surface application of manure to field plots

Transfer of phosphorus (P) from surface-applied manures to runoff is an important source of pollution, but few studies have closely monitored P dynamics in manure, soil, and runoff through time. We monitored manure and soil P over 14 to 17 months in field experiments in Texas and Pennsylvania, USA following dairy and poultry manure surface application. Manure was applied to porous fabric that enabled discrete sampling of both manure and underlying soil. Manure mass consistently decreased while manure total P was essentially constant through time. Manure water extractable P decreased rapidly for the first two months, likely due to rainfall leaching, but then maintained stable concentrations thereafter, with other forms of manure P gradually transformed to water extractable forms. Soil P from the upper 2 cm rapidly increased after manure application in association with manure leaching by rain. After 2 to 3 months, soil P peaked and either remained constant or gradually declined. Similar trends occurred at 2–5 and 5–10 cm, but with lesser magnitudes. At 10–15 cm, soil P changed little over time. In Pennsylvania, naturally occurring runoff from 0.7-m × 1.3-m plots without and without manure was also monitored. Runoff dissolved P concentrations were greatest for the first event after manure application and decreased steadily through time, but remained greater than P concentrations from control plots, and were always well related to manure water extractable P. This study reveals that management practices for water quality protection must consider the potential for manure P transformations to contribute dissolved P to runoff long after manure is applied.     

Application of manure to no-till soils: phosphorus losses by sub-surface and surface pathways

The acceleration of surface water eutrophication attributed to agricultural runoff has focused attention on manure management in no-till. We evaluated losses of phosphorus (P) in sub-surface and surface flow as a function of dairy manure application to no-till soils in north-central Pennsylvania. Monitoring of a perennial spring over 36 months revealed that dissolved reactive P (DRP) concentrations increased 3- to 28-fold above background levels whenever manure was broadcast to nearby field soils. A study conducted with 30-cm deep intact soil cores indicated that incorporation of manure by tillage lowered P loss in leachate relative to broadcast application, presumably due to the destruction of preferential flow pathways. More P was leached from a sandy loam than a clay loam soil, although differences between soils were not as great as differences between application methods. In contrast, rainfall simulations on 2-m² field runoff plots showed that total P (TP) losses in surface runoff differed significantly by soil but not by application method. Forms of P in surface runoff did change with application method, with DRP accounting for 87 and 24% of TP from broadcast and tilled treatments, respectively. Losses of TP in leachate from manured columns over 7 weeks (0.22–0.38 kg P ha⁻¹) were considerably lower than losses in surface runoff from manured plots subjected to a single simulated rainfall event (0.31–2.07 kg TP ha⁻¹). Results confirm the near-term benefits of incorporating manure by tillage to protect groundwater quality, but suggest that for surface water quality, avoiding soils prone to runoff is more important.

Fertilizer nitrogen balance study on sandy loam with winter wheat

The uptake of labelled nitrate was studied under field conditions using a sandy loam soil and winter wheat as the crop. Thirty per cent of the added labelled nitrate was found in the grain, 13% in the straw plus chaff and only 2% in the harvested roots. The labelled nitrogen in the plant comprised about 6% of labelled fertilizer origin, and about 94% from the soil. By analysing the soil after harvest, it was shown that 16% of the fertilizer nitrogen was in the 0–6 cm layer, 19% in the 6–12 cm layer, 6% in the 12–25 cm layer and 5% in the 25–50 cm layer. This means a total residual fertilizer nitrogen of 46%. A study was made of the border effect by analysing the plants of the first and second row around the fertilized plot. The results show that the plants of the first row took up 1.4% of the added fertilizer and those of the second row only 0.4%. The total recovery of the added fertilizer in soil and plant was 93.1%, indicating that 6.9% was lost under conventional cultivation practice.     

Fertilizer P transformations and P availability in hillslope soils of northern Ghana

Alfisols of the Savannas in northern Ghana have high base saturation and moderate P sorption capacities. Lateritic nodules are common, occurring in highly variable quantities at different landscape positions. Such nodules can have high P sorption capacities, and therefore effectiveness of P fertilisation may depend on landscape position. The objective of this study was to investigate the effectiveness of Togo rock phosphate (TRP), 50% acidulated TRP (PAPR) and single super phosphate (SSP) in providing and maintaining available P.Phosphorus supplying ability of soils from upper and lower slopes of three locations in northern Ghana was studied with and without fertilizer addition by repeated desorption with anion exchange resin membrane (AEM) burial in the laboratory and by measuring dry matter yield (DMY) and P uptake of sorghum for six successive greenhouse croppings. Transformations of the applied fertilizers were studied by sequential extraction.Phosphate desorbed with AEM after 38 days, the DMY and the P uptake of sorghum all followed the order SSP > PAPR > TRP = control. The relative agronomic effectiveness of the PAPR was 63% of SSP. Although half the applied TRP was transformed to other forms than acid extractable apatite, this did not liberate P to the AEM. Less than 10% of the SSP was absorbed by the AEM.In the nodule-rich upper slope soils, initial availability of added P was higher, but decreased more rapidly than in lower slope soils. We attributed this to an initial concentration of fertilizer in the smaller volume of soil fines followed by slow sorption into the nodules. At the upper slope, much of the TRP and PAPR were transformed to less available forms, while at the lower slope more TRP remained as untransformed apatite.     

Effects of nitrification inhibitors and time and rate of slurry and fertilizer N application on silage maize yield and losses to the environment

Field experiments with silage maize during eight years on a sandy soil in The Netherlands, showed that dicyandiamide (DCD) addition to autumn-applied cattle slurry retarded nitrification, thus reducing nitrate losses during winter. Spring-applied slurry without DCD, however, was on average associated with even lower losses and higher maize dry matter yields.Economically optimum supplies of mineral N in the upper 0.6 m soil layer in spring (EOSMN), amounted to 130–220 kg ha^–1. Year to year variation of EOSMN could not be attributed to crop demand only. According to balance sheet calculations on control plots, apparent N mineralization between years varied from 0.36 to 0.94 kg ha^–1 d^–1. On average, forty percent of the soil mineral N (SMN) supply in spring, was lost during the growing season. Hence, the amounts of residual soil mineral N (RSMN) were lower than expected. Multiple regression with SMN in spring, N crop uptake and cumulative rainfall as explanatory variables, could account for 79 percent of the variation in RSMN.Postponement of slurry applications to spring and limiting N inputs to economically optimum rates, were insufficient measures to keep the nitrate concentration in groundwater below the EC level for drinking water.     

Long term effects of inorganic fertilizer inputs on crop productivity in a rice-wheat cropping system

Three levels of N (40,80,120 kg N ha^-1) and P (0,17.5,35 kg P ha^-1), and 2 levels of K (0,33 kg K ha^-1) were tested for 19 years in rice and wheat crops of a rice-wheat cropping system in a fixed layout of 3×2×2 factorial partially confounded design along with one control and 3 replications. From this trial, data of 7 treatments, i.e. 0-0-0, 40-35-33, 80-35-33, 120-35-33, 120-0-0, 120-35-0 and 120-0-33 kg ha^-1 N-P-K respectively were compared for yield trends, changes in response functions, soil organic -C and available N,P,K status. Soil organic - C decreased in unfertilized plots by 62% (over initial value of 0.45%) but increased by 44, 40 and 36% in plots receiving 120-35-33, 120-35-0 and 80-35-33 kg ha^-1 N-P-K respectively. Available N was also greatest in these same three treatments. Available soil P increased by about 5 fold in 15 years in treatments supplied with fertilizer P, but no significant change was detected in treatments without P addition. Yields of rice and wheat exhibited linear declining trend in all treatments. The highest rate of decline (89 kg ha^-1 year^-1 in rice and 175 kg ha^-1 year^-1 in wheat), however, was found when 120 kg ha^-1 N was applied alone. The least rate of decline of 20 kg ha^-1 year^-1 in rice and 58 kg ha^-1 year^-1 in wheat was observed when 40-35-33 kg ha^-1 N-P-K respectively was applied to both the crops. At currently recommended levels of NPK (120-35-33 kg ha^-1), the rate of decline in yields was 25 kg ha^-1 year^-1 for rice and 62 kg ha^-1 year^-1 for wheat. Possible causes of these yield declines are discussed.