HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Figures Only Full Text (PDF) Free Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in ISI Web of Science Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via ISI Web of Science (16) Citing Articles via Google Scholar Google Scholar Articles by Ladha, J.K. Articles by Watanabe, I. Search for Related Content PubMed Articles by Ladha, J.K. Articles by Watanabe, I. Agricola Articles by Ladha, J.K. Articles by Watanabe, I.

DIVISION S-3-SOIL BIOLOGY & BIOCHEMISTRY

Long-Term Effects of Urea and Green Manure on Rice Yields and Nitrogen Balance

^a W. Ventura, Crop, Soil and Water Sciences Division, IRRI, MCPO Box 3127, 1271 Makati City, Philippines
^b Social Sciences Division, IRRI, Muscle Shoals, AL USA
^c International Fertilizer Development Center, P.O. Box 2040, Muscle Shoals, AL USA
^d 6-58-18, Jindaiji-Kita, Chofu-shi, Tokyo, 182-0011, Japan

j.k.ladha{at}cgiar.org

	ABSTRACT

TOP ABSTRACT INTRODUCTION Materials and methods Results and discussion Conclusions REFERENCES

 
Data from a 14-yr double-crop rice (Oryza sativa L.) experiment allowed comparison of the long-term effects of N fertilizer from different sources (urea and in situ grown azolla [Azolla microphylla Kaulf.] and sesbania [Sesbania rostrata Bremek. & Oberm.]) on N balances, soil N pools (both total and available), and yields. Although data show that plant-available N was maintained over time in both wet (WS) and dry seasons (DS), yields declined significantly, indicating a decline in physiological N use efficiency. The yield declines were generally similar regardless of N source in both seasons. The WS decline averaged 150 kg ha^-1 yr^-1 in the three added-N treatments, while the DS decline averaged 185 kg ha^-1 yr^-1. After 27 crops, the cumulative positive N balance was estimated at 1244, 348, 646, and 1039 kg N ha^-1 in control, urea, sesbania, and azolla treatments, respectively. There was no significant change in soil total N content in the control and urea treatments, whereas it increased to 344 to 541 kg after 27 crops in the sesbania and azolla treatments. Conservation of the soil N status and positive N balance, in spite of the high amounts of N removed through grain and straw, reflect the N contribution (13–46 kg ha^-1 crop^-1) from nonsymbiotic N₂ fixation. In addition, sesbania and azolla were estimated to add 57 to 64 kg ha^-1 crop^-1 through symbiotic N₂ fixation. These results demonstrate that in rice–rice cropping systems biological N fixation plays a vital role in replenishing the soil N pool. However, continuous application of green manure N (GM-N) did not increase crop N availability, perhaps because of the presence of a recalcitrant soil organic matter fraction. Residual effects on rice grain yield and N uptake were observed only with GM-N sources.

Abbreviations: BNF, biological N fixation • DS, dry season • GM, green manure • GM-N, green manure N • WS, wet season

	INTRODUCTION

TOP ABSTRACT INTRODUCTION Materials and methods Results and discussion Conclusions REFERENCES

 
RICE IS THE WORLD'S most important staple food for more than two billion people in Asia and for hundreds of millions in Africa and Latin America. Subsistence rice farming of the pre-chemical era sustained the N status of soils by maintaining an equilibrium between N loss from crop harvest and N gain from biological N fixation (BNF) (Ladha and Peoples, 1995). In today's intensive rice monocropping systems, however, this equilibrium has been disturbed, with inputs from mineral fertilizers now playing a significant role.

The objectives of this study were to examine the long-term effects of urea applications on N balances, soil N pools (both total and available), and yields, and to ascertain the effects when green manure (GM) grown in situ (azolla and sesbania) was substituted for urea. The analysis is based on a 14-yr double-crop rice experiment conducted at the International Rice Research Institute. Long-term experiments are essential for accurately measuring nutrient balances because large year-to-year variation in crop growth can dominate measurements taken during short periods and because the changes must be measured against the large quantities of nutrient usually present in the soil (Greenland, 1994; Powlson, 1994).

Yield trends have been estimated in many other long-term experiments, but it is rare for these experiments to conduct periodic monitoring of changes in N uptake and soil N pools. Such measurements can provide valuable insight into the possible causes of observed yield trends and the implications of these trends for sustainability. In addition, most tropical experiments were designed to study yield trends with mineral N input, while ignoring GM as an alternative source of N. There has been much speculation on the long-term effects of continuous N use, particularly GM-N, but actual data for intensive rice systems are lacking (Bouldin, 1988).

	Materials and methods

TOP ABSTRACT INTRODUCTION Materials and methods Results and discussion Conclusions REFERENCES

 
Experimental Details
A long-term experiment began in March 1985 in an isohyperthermic Andaqueptic Haplaquoll (Maahas silty clay) at the experimental farm of the International Rice Research Institute, Los Baños, Philippines, with the following soil characteristics: pH 6.3 (1:1 w/v water), 12.5 g organic C kg^-1, 1.4 g total N kg^-1, 12.3 mg Olsen P kg^-1, 0.43 cmol_c K kg^-1, and cation-exchange capacity of 42.3 cmol_c kg^-1 in the top 20-cm layer. Two crops of rice, one during the WS (transplanting, May–August, and harvest, August–December) and the second during the dry season (DS) (transplanting, October–February, and harvest, February–June), were grown with full irrigation every year with four treatments and 7 by 7 m plot size: no applied N (control, Treatment 1), mineral N as urea (Treatment 2), sesbania (Treatment 3), and azolla (Treatment 4). Except for the control, which had three replicates, the treatments were replicated four times in a randomized complete block design (unbalanced). Treatment 1 received no N from any source. In Treatment 2, urea was applied at 50 kg N ha^-1 to Crops 1 to 3, 60 kg N ha^-1 to Crops 4 to 14, and 80 kg N ha^-1 to Crops 15 to 27. Urea was split-applied, with one-half basally incorporated and the other half broadcast before panicle initiation. In Treatment 3, azolla was inoculated at 0.3 kg fresh weight m^-2 after the first plowing and harrowing (29–43 d before transplanting of rice). Azolla microphylla 4018 was used in Crops 1 to 7 and 10 to 27, and a mixed inoculum of A. microphylla 4018 and a hybrid strain 4028 (A. microphylla 4018 and A. filiculoides 1001) (Watanabe et al., 1992) was used in Crops 8 to 9. The growing period of azolla varied from 42 to 62 d (June–August during WS and December–February during DS), and azolla was incorporated into the soil three to four times during each rice crop (2–3 times before transplanting and once 20 to 30 d after transplanting). In Treatment 4, sesbania or aeschynomene (Sesbania rostrata 24062 in all the crops except for Aeschynomene afraspera J. Leonard 14054 in Crop 8) seed was broadcast at the rate of 30 to 40 kg ha^-1 after plowing and draining on the wet field. The growing period of sesbania varied from 46 to 67 d (June–August during WS and December–February during DS).

The plots were irrigated carefully to avoid contamination among plots and were kept submerged (3–5 cm water layer) during rice and sesbania or azolla growth. The plots were drained 15 d before the rice harvest and were reflooded and wet-tilled 7 d after harvest for sowing of sesbania and inoculation of azolla. Plots were drained 3 d before incorporation of green manure. Sesbania was cut at the soil level and chopped into 20- to 30-cm pieces. Puddling was done with a hydrotiller, which also incorporated the green manure.

One day after the incorporation of GM, rice seedlings (15–29 d old) were transplanted to give a 20 by 20 cm and 10 by 20 cm spacing for Crops 1 through 20 and Crops 21 through 27, respectively. Rice cultivars IR54, IR64, IR66, and IR68 were used for Crops 1 through 4, 5, 6, and 7, respectively. IR72 was used for Crops 8 through 27. Pest and disease control measures in sesbania, azolla, and rice were used as necessary.

The azolla treatment was the only treatment receiving P before Crop 14. The azolla treatment received a total of 5 to 10 kg water-soluble P ha^-1 every season during azolla growth. Starting with Crop 15, azolla and sesbania received 20 kg P ha^-1, but rice received none, whereas in the control and urea-N plots 30 kg P ha^-1 was applied to each rice crop. All plots received P as solophos except azolla, which received P as triple superphosphate. In addition, starting with Crop 15, each crop of rice in all treatments received 30 kg K ha^-1as muriate of potash.

Residual Effects
The residual effects of the N source treatments were assessed in Crops 10 (DS, 1990) and 17 (WS, 1993). In these two crops, N source treatments were not applied so that the rice was completely dependent on the indigenous soil N supply. Crop and soil management were otherwise similar to those of other crop cycles. During the fallow period of Crops 10 and 17, the plots were kept with a 3- to 5-cm water layer.

Soil and Plant Sampling and Analysis
Soil samples from the 0- to 25- and 26- to 50-cm depths from three sites in each plot were taken using a bottomless 20 by 20 by 65 cm (width by length by depth) core sampler at 15 d after the rice harvest of Crops 1, 7, 16, and 20, and a tube of 4.7-cm i.d. for Crop 27, for total N and for bulk density determinations. The entire volume of soil was weighed and mixed thoroughly and a subsample was taken from the mixed soil. The dry weight of soil in this sample was determined. The wet soils were sieved through a 1-mm screen and mixed, and two composite samples were taken from each plot. Any plant material remaining on the screen was assayed for total N and included in plant N. The wet soil samples were sun-dried and analyzed for total N by the macro-Kjeldahl method (Bremner and Shaw, 1958). A standard soil sample was included in every batch of samples to make corrections.

Azolla biomass was estimated before incorporation by collecting samples in a 25 by 25 cm metal frame from 10 locations in each plot. Sesbania biomass was determined from two 1-m² subsamples in each plot. Rice yield was taken from 8 m² at opposite ends of the plot, starting from the fifth row. The plants were cut 3 cm from the ground. Grain and straw were separated in a rice threshing machine, dried in a batch drier, and weighed. Grain water content was determined immediately after weighing and subsamples were dried again in an oven at 65°C for dry weight determination. Grain weight was expressed at 140 g kg^-1 water content. Aboveground dry weight of rice (grain plus straw), azolla, and sesbania was determined after oven drying at 70°C for 72 h. Nitrogen in biomass of rice (grain plus straw), azolla, and sesbania was determined by the micro-Kjeldahl method (Yoshida et al., 1972).

Data Analysis
Linear regression analyses of the form were estimated using the LIMDEP statistical package in order to determine the magnitude of yield trends after controlling for the effects of changing weather and N input during the course of the experiment. Y_t is the grain yield in year t (kg ha^-1), NI_t is the level of N input in year t (kg N ha^-1), SR_t is the average solar radiation (KJ m^-2 d^-1) in the 30 d before harvest in year t, T is a time trend variable, and the a_is are the regression coefficients to be estimated. Average daily solar radiation in the 30 d before harvest was used because solar radiation during the reproductive and ripening stages has the greatest effect on rice grain yield (Yoshida, 1981). Regressions for wet and dry season yields were estimated separately because the underlying biophysical processes are different in the two seasons. Separate regression equations were estimated for each of the four treatments. Regressions were estimated both including and excluding Crops 10 and 17, the two crops for which no N was applied in any of the four treatments. The results were not sensitive to inclusion or exclusion of these two crops, and only the results including all data are reported.

Simulation models provide an alternative to regression analysis for taking account of changing weather conditions and N applications. Thus, the CERES rice model (Singh et al., 1993) was used to simulate yields for each of the four treatments in both seasons of each year in order to test the robustness of the statistical techniques. Each simulation run used the transplanting date and applied N rate particular to each crop as well as the weather data for each particular year and season (solar radiation, temperature, and rainfall).

To provide insight into the possible mechanisms underlying the estimated yield trends, two additional sets of regressions were also estimated. The first set of regressions (separate regressions for each treatment and season) was of the form , where all variables are as above and NU is N uptake in year t (kg ha^-1 crop^-1). The second set of regressions was of the form . These additional regressions allow for a decomposition of the yield trend into (i) trends in N uptake with time and (ii) trends in yield after controlling for changing levels of N uptake with time. The coefficient b₃ measures changes in the plant's ability over time to recover N from soil and fertilizer, and the coefficient c₃ measures changes in the plant's ability over time to convert any given level of N uptake into grain yield. The effect on yield of changes over time in the plant's ability to recover N is calculated as the product of the coefficients b₃ (the annual rate of decline in N uptake) and c₁ (the effect of N uptake on yield). The effect on yield of changes over time in the ability of the plant to convert N uptake into grain yield is measured directly as the coefficient c₃. Standard measures of recovery efficiency (increase in plant N per unit of applied N) and physiological efficiency (grain yield increase per unit increase in plant N) were also calculated for illustrative purposes. However, the meaning of these measures is less clear in an experiment where the level of N input is changing by large amounts from year to year.

A N balance sheet was constructed considering the different inputs (urea and GM-N, BNF, irrigation, and rain N) and outputs (crop harvest; grain plus aboveground plant material). The roots and stubble (3 cm from the ground) biomass that remained in the soil was not quantified. Nitrogen from irrigation water and pesticides was determined. Nitrogen input from nonsymbiotic N fixation in soil was estimated from total soil N balance based on Kjeldahl data (App et al., 1980). Nitrogen input by rainfall was based on measurements made earlier (App et al., 1984). Analyses of variance were done to test whether N balances in the different treatments were significantly different from zero. Duncan's multiple range test was used to test differences between treatments (SAS Institute, 1995).

	Results and discussion

TOP ABSTRACT INTRODUCTION Materials and methods Results and discussion Conclusions REFERENCES

 
Green Manure N Production, Rice Yield and N Uptake, and N Use Efficiency
On an average basis, N inputs from the three applied sources ranged from 69 to 75 kg ha^-1 in the DS and 68 to 88 kg ha^-1 in the WS (Table 1) . The amount of biomass and N of sesbania and azolla varied greatly, largely because of differences in growing period and seasons. Sesbania, a tall stem-nodulating legume, tends to produce more biomass than azolla, which is a small fern that floats on the soil–water surface.

View this table: [in this window] [in a new window]   Table 2 Averages of rice grain yield, N uptake, N-recovery efficiency, agronomic efficiency, and physiological efficiency in 25 crops, IRRI field, 1985 to 1998.

View this table: [in this window] [in a new window]   Table 3 Significance of F values in the combined analysis of variance for rice yield, N uptake, N recovery efficiency (RE), agronomic efficiency (AE), and physiological efficiency (PE) in 27 crops (1985–1998).

Trends of Rice Yield and N Uptake
Linear regressions of yield on time, N input, and solar radiation showed higher rice yields with increasing levels of N, as indicated by the positive coefficient on the N input variable (Table 4) . The coefficients on N input were statistically significant (P < 0.01) in the DS in all treatments, but only the coefficient on N input in the urea treatment was statistically significant (P < 0.05) in the WS. The coefficients on solar radiation were positive in nearly all cases, but were statistically significant (P < 0.01) only in the azolla and sesbania treatments in the DS.

View larger version (25K): [in this window] [in a new window]   Fig. 1 Rice yield trends in the double-crop system of the long-term biofertilizer experiments conducted at the IRRI research farm. Since N input and solar radiation changed during the experiment, the regression lines are plotted holding N input and solar radiation constant over time at their mean levels in each treatment. See Table 4 for regression coefficients

In the WS, the results of the decomposition analysis showed that the coefficient on the time variable (b₃) in the regression of N uptake on N input, SR, and time was negative but not statistically significant in any of the four treatments (Table 5) . However, in the regression of yield on N uptake, SR, and time, the coefficient on the time variable (c₃) was negative and statistically significant (P < 0.05) in all four treatments, indicating a decline in the ability of the plant to convert plant N to grain yield (Table 6) . The product of the b₃ and c₁ coefficients shows that a declining trend in N uptake with time accounted for only a small percentage of the yield decline as measured by the regression coefficients in Table 6 (19% on average across the four treatments). The c₃ coefficient shows that a declining ability to convert plant N to grain yield accounted for a much larger share of the yield decline (82% on average across the four treatments). (The two figures do not add up to one because they are the results of statistical estimates.) Calculations of recovery efficiency and physiological efficiency showed similar results. Comparing the last 3 yr of the experiment to the first 3 yr, physiological efficiency declined by 52, 49, and 32% in the urea, azolla, and sesbania treatments, respectively. On the other hand, N uptake efficiency increased by 30 and 6% in the urea and azolla treatments and declined by 29% in the sesbania treatment. These results are consistent with those from the regression analysis. For the DS, the same set of decomposition regressions again showed that a declining ability to convert plant N to grain yield was the dominant effect, but the coefficients on the time trend variables in the regressions were not statistically significant (results not shown).

Cassman et al. (1995) hypothesized that yield declines in other long-term experiments conducted at IRRI were due to a decline in soil N-supplying capacity and the ability of the rice plant to recover N resulting from changes in the composition of organic matter. The data from this experiment are not consistent with that hypothesis, but the experiments cited by Cassman et al. (1995) began 20 yr earlier than the experiment analyzed here and may not be comparable. Long-term data on N uptake were not collected in these other experiments, however, so Cassman et al. (1995) could not test their hypothesis using the methods in this paper.

The fact that grain yield declined in this experiment without a decline in crop N uptake may indicate that N was not available to the crop at critical growth phases. This suggests another hypothesis: that there is a change in the pattern of soil N mineralization and availability to the crop. This is likely to happen when a soil remains continuously submerged and frequently puddled with incorporation of crop residue (root and stubble) of high C/N ratio (Kundu and Ladha, 1995). Continuous flooding and repeated puddling of the soil without thorough dry tillage drastically reduces the water percolation rate, often to about zero. Intensive reduction and anaerobic decomposition of organic material lead to a large accumulation of reducing substances in the soils, and this may strongly change the pattern of soil N mineralization. This hypothesis is consistent with our data on total soil N, which was either maintained or increased with time. However, further work examining N mineralization patterns in relation to soil water regimes, puddling, and addition of root and stubble is needed, since as yet no hard data support such a hypothesis.

Nitrogen Balance
The cumulative N balance after Crops 7, 16, 20, and 27 reported in Table 7 is positive and statistically different from zero in all treatments. There was no difference in positive N balance or N gain among treatments after Crop 7. The N gain in the control plots where no N was added ranged from 44 to 48 kg ha^-1 crop^-1 and remained unchanged with cropping cycle. The average N gains per crop in the sesbania and azolla plots were 62 and 79 kg ha^-1 crop^-1 after Crop 7 but declined gradually to 24 and 38 ha^-1 crop^-1, respectively, after Crop 27. The positive N balance of the azolla-treated plots after 16 crops was significantly higher than that of the control and the other treatments. However, after Crops 20 and 27, the N balance decreased and was not significantly different from that of the control. The N balances of the sesbania-treated plots were not significantly different from that of the control after Crops 16 and 20, but were significantly lower than that of the control after Crop 27. The N gain per crop decreased from 43 kg ha^-1 crop^-1 after Crop 16 to 24 kg ha^-1 crop^-1 after Crop 27 in this treatment. The application of P fertilizer to correct P deficiency in the sesbania plots starting from Crop 14 did not improve the N gain (Ventura and Ladha, 1997). The N gain with urea-N was the lowest and also declined with time. Although N gain showed a declining trend with time in all the treatments except the control, total soil N was higher in azolla and sesbania and remained unchanged in the control and urea-N treatments (Fig. 2) . In the azolla and sesbania treatments, soil N increased up to Crop 16, but appeared to establish equilibrium thereafter. This indicates the positive effect of the continuous application of azolla and sesbania in building soil N.

View larger version (19K): [in this window] [in a new window]   Fig. 2 Total soil N of the long-term biofertilizer plots after 7, 16, 20, and 27 rice crops

Because all the possible N inputs (from fertilizer, sesbania or azolla, rainwater, and irrigation water) and outputs (crop removal and change in soil pool) were considered, the N gains are probably due to nonsymbiotic N₂ fixation (App et al., 1980). In rice, the amount of N₂ fixation is much lower than that of a legume and most of the N₂ fixation occurs in soil. Therefore, a change in N balance is the most accurate measure for estimating N₂ fixation. It appears that at least the equivalent of 75% of the N uptake by the 27 crops with no addition of N was derived from nonsymbiotic N₂ fixation. When rice was grown continuously with urea-N, the nonsymbiotic N₂ fixation decreased to 13% of crop N uptake. On the other hand, the contribution of nonsymbiotic N₂ fixation with an organic source of N ranged from 25 to 40%. Because the N balance figures are the sum of measured inputs and outputs, however, and because N losses were not measured, the contribution of actual N₂ fixation is underestimated. Assuming a 30% N loss (a conservative estimate) from urea and organic sources of N, the contributions of nonsymbiotic N₂ fixation of total plant uptake would be 33, 58, and 46% with urea, azolla, and sesbania N, respectively. It is likely, however, that the contribution of nonsymbiotic N₂ fixation in sesbania and azolla plots is slightly overestimated because symbiotically fixed N might have been provided as sloughed off leaves and roots in the azolla plot and as residues and sloughed off roots in the sesbania plot. Nevertheless, this study clearly shows that nonsymbiotic N₂ fixation plays a vital role in replenishing the soil N pool, thereby sustaining rice yields in both less intensive production systems with no input of N and highly intensive production systems with a high input of N.

In addition, it is also important to note that sesbania and azolla derived most of their N from symbiotic associations with rhizobium and anabaena, respectively. Assuming 80% N derived from the atmosphere, sesbania (Ladha et al., 1992) and azolla (Watanabe et al., 1991) added 1600 and 1426 kg N ha^-1, respectively, after 27 crops.

Residual Effects of Green Manure and Urea-N on Rice Yield and N Uptake
No N or GM was applied in Crops 10 and 17. The no-N control and urea-N treatments failed to show any residual effects on rice grain yield and rice N uptake, which is consistent with insignificant changes in total soil N. However, the continuous incorporation of azolla or sesbania for 11 to 16 crops, resulted in residual effects equivalent to about 0.5 Mg ha^-1 grain yield and 10 kg ha^-1 N uptake, which were significantly different in both Crops 10 and 17 from the no-N and urea-N treatments. The residual response of both inorganic and organic sources of N to grain yield and N uptake remained unchanged from Crop 10 to Crop 17, indicating no effect of additional applications of GM or urea-N. The residual effects of 10 kg N uptake ha^-1 are less than 3% of the 390 and 817 kg N ha^-1 that were accumulated in the soil by continuous use of sesbania and azolla after 16 crops. Becker et al. (1994) found no significant effect of continuous incorporation of sesbania for two crops on a subsequent third rice crop. In an upland system based on ¹⁵N, Jensen (1994) found only 1 to 2% of the peas' residual N mineralized after 2 yr of decomposition. It is often assumed that the GM-N not used by the succeeding crop after its application will be available to subsequent crops, and that the cumulative effects of continued use of GM are important, not only for N supply but also for soil productivity (Bouldin, 1988). Long-term data on recovery of GM-N were lacking, however, when this assertion was made. This study shows that continuous application of GM increased total soil N, but that its availability to the plant was limited. The remaining residual N is probably present in a recalcitrant soil organic matter fraction that decomposes slowly over a long period.

	Conclusions

TOP ABSTRACT INTRODUCTION Materials and methods Results and discussion Conclusions REFERENCES

 
The results demonstrate that, in a lowland continuous double-rice cropping system, both total and plant-available soil N pools are maintained on a long-term basis whether urea or GM is the source of N. After 27 crops, the cumulative positive N balance was estimated at 1244, 348, 646, and 1039 kg N ha^-1 in control, urea, sesbania, and azolla treatments, respectively. There was no change in total soil N content in the control and urea treatments, while after 27 crops it increased to 344 kg ha^-1 in the sesbania treatment and 541 kg ha^-1 in the azolla treatment. Because all the inputs and outputs (except N losses) were considered, these positive N balances were attributed to nonsymbiotic N₂ fixation by associative and free-living microorganisms. The positive N balances ranged from 13 to 48 kg ha^-1 crop^-1 after 27 continuous crops. The estimated levels of N₂ fixation would be higher if N losses were taken into account. These results clearly demonstrate that in rice–rice systems biological N₂ fixation plays a vital role in replenishing the soil N pool by helping to maintain a good balance between N losses and N gains (Ladha, 1998). After controlling for variable N input, the plant-available N (crop uptake) was maintained over time in both the wet and dry seasons. Rice yields declined with time, however, indicating a decline in physiological N use efficiency.

	ACKNOWLEDGMENTS

 
We acknowledge Ms. Agnes Tirol-Padre for helping us in chemical analysis, Dr. Abedullah for assistance in performing the regression analysis, and Dr. S. Peng for constructive comments on the manuscript.

Received for publication July 15, 1999.

	REFERENCES

TOP ABSTRACT INTRODUCTION Materials and methods Results and discussion Conclusions REFERENCES

 

• App A., Santiago T., Daez C., Menguito C., Ventura W., Tirol A., Po J., Watanabe I., De Datta S.K., Roger P. Estimation of the nitrogen balance for irrigated rice and the contribution of phototrophic nitrogen fixation. Field Crops Res. 1984;9:17-27.
• App A., Watanabe I., Alexander M., Ventura W., Daez C., Santiago T., De Datta S.K. Nonsymbiotic nitrogen fixation associated with the rice plant in flooded soils. Soil Sci. 1980;130:283-289.
• Becker M., Ladha J.K., Simpson I.C., Ottow J.C.G. Parameters affecting residue nitrogen mineralization in flooded soils. Soil Sci. Soc. Am. J. 1994;58:1666-1671.[Abstract/Free Full Text]
• Bouldin, D.R. 1988. Effect of green manure on soil organic matter content and nitrogen availability. p. 151–163. In Proceedings of a symposium on sustainable agriculture: Green manure in rice farming. IRRI, Manila, Philippines.
• Bremner J.M., Shaw K. Denitrification in soil: 1. Methods of investigation. J. Agric. Sci. 1958;51:22-39.
• Cassman K.G., De Datta S.K., Olk D.C., Alcantara J., Samson M., Descalsota J., Dizon M. Yield decline and the nitrogen economy of long-term experiments on continuous, irrigated rice systems in the tropics. In: Lal R., Stewart B.A., eds. Soil management: Experimental basis for sustainability and environmental quality. CRC Press, Boca Raton, FL: Lewis Publ, 1995:181-222.
• Dawe D., Dobermann A., Moya P., Abdulrachman S., Bijay-Singh, Lal P., Li S.Y., Lin B., Panaullah G., Sariam O., Singh Y., Swarup A., Tan P.S., Zhen Q.X. How widespread are yield declines in long-term rice experiments in Asia?. Field Crops Res. 2000;66:175-193.
• Dobermann, A., and T. Fairhurst. 2000. Nutrient disorders and nutrient management. Potash and Phosphate Inst., Potash and Phosphate Inst. of Canada, and IRRI, Singapore and Los Banos. (In press).
• Flinn J.C., De Datta S.K. Trends in irrigated-rice yields under intensive cropping at Philippine research stations. Field Crops Res. 1984;9:1-15.
• Greenland D.J. Long-term cropping experiments in developing countries: The need, the history and the future. In: Leigh R.A., Johnston A.E., eds. Long-term experiments in agricultural and ecological sciences. Wallingford, UK: CAB Int, 1994:187-209.
• Jensen S.E. Availability of nitrogen in ¹⁵N-labelled mature pea residues to subsequent crops in the field. Soil Biol. Biochem. 1994;26:465-472.
• Kundu D.K., Ladha J.K. Efficient management of soil and biologically fixed N₂ in intensive-cultivated rice fields. Soil Biol. Biochem. 1995;27:431-439.
• Ladha J.K. Role of biological nitrogen fixation in replenishing soil nitrogen pool in cropping systems. In: Elmerich C., et al. , ed. Biological nitrogen fixation for the 21st century. Dordrecht, the Netherlands: Kluwer Academic Publ, 1998:605-608.
• Ladha J.K., Peoples M.B. Management of biological nitrogen fixation for the development of more productive and sustainable agricultural systems: Preface. Plant Soil 1995;174:1.
• Ladha J.K., Pareek R.P., Becker M. Stem-nodulating legume-Rhizobium symbiosis and its agronomic use in lowland rice. Adv. Soil Sci. 1992;20:148-192.
• Powlson D.S. Quantification of nitrogen cycles using long-term experiments. In: Leigh R.A., Johnston A.E., eds. Long-term experiments in agricultural and ecological sciences. Wallingford, UK: CAB Int, 1994:97-115.
• Roger P.A., Ladha J.K. Biological N₂ fixation in wetland rice fields: Estimation and contribution to nitrogen balance. Plant Soil 1992;141:41-55.
• SAS Institute. 1995. SAS system for Windows. Release 6.11. SAS Inst. Cary, NC.
• Singh, U., J.T. Ritchie, and D.G. Godwin. 1993. A user's guide to CERES-Rice. V2.10. Simulation manual EFDC-SM-4. Int. Fert. Dev. Center, Muscle Shoals, AL.
• Ventura W., Ladha J.K. Sesbania phosphorous requirements when used as biofertilizer for long-term rice cultivation. Soil Sci. Soc. Am. J. 1997;61:1240-1244.[Abstract/Free Full Text]
• Watanabe I., Roger P.A., Ladha J.K., Van Hove C. Biofertilizer germplasm collection at IRRI. Los Baños, Philippines: IRRI, 1992.
• Watanabe I., Yoneyama T., Talukda H., Ventura W. The contribution of atmospheric N₂ to Azolla spp. grown on flooded soil. Soil Sci. Plant Nutr. 1991;37:101-109.
• Witt C., Cassman K.G., Olk D.C., Biker U., Libson S.P., Samson M.I., Ottow J.C.G. Initial effects of crop rotation and residue management on carbon sequestration, nitrogen cycling and crop productivity of irrigated rice systems. Plant Soil 2000;in press.
• Yoshida S. Fundamentals of rice crop science. Manila, Philippines: IRRI, 1981.
• Yoshida, S., D. Forno, J. Cock, and K. Gomez. 1972. p. 11–12. In Laboratory manual for physiological studies of rice. IRRI, Manila, Philippines.

This article has been cited by other articles:

				M. F. Pampolino, E. V. Laureles, H. C. Gines, and R. J. Buresh Soil Carbon and Nitrogen Changes in Long-Term Continuous Lowland Rice Cropping Soil Sci. Soc. Am. J., May 1, 2008; 72(3): 798 - 807. [Abstract] [Full Text] [PDF]

				C. M. Cherr, J. M. S. Scholberg, and R. McSorley Green Manure Approaches to Crop Production: A Synthesis Agron. J., February 7, 2006; 98(2): 302 - 319. [Abstract] [Full Text] [PDF]

				A. Tirol-Padre and J. K. Ladha Assessing the Reliability of Permanganate-Oxidizable Carbon as an Index of Soil Labile Carbon Soil Sci. Soc. Am. J., May 1, 2004; 68(3): 969 - 978. [Abstract] [Full Text] [PDF]

				A. Dobermann, C. Witt, S. Abdulrachman, H. C. Gines, R. Nagarajan, T. T. Son, P. S. Tan, G. H. Wang, N. V. Chien, V. T. K. Thoa, et al. Soil Fertility and Indigenous Nutrient Supply in Irrigated Rice Domains of Asia Agron. J., July 1, 2003; 95(4): 913 - 923. [Abstract] [Full Text] [PDF]

This Article Abstract Figures Only Full Text (PDF) Free Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in ISI Web of Science Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via ISI Web of Science (16) Citing Articles via Google Scholar Google Scholar Articles by Ladha, J.K. Articles by Watanabe, I. Search for Related Content PubMed Articles by Ladha, J.K. Articles by Watanabe, I. Agricola Articles by Ladha, J.K. Articles by Watanabe, I.