Agronomy Journal 92:875-879 (2000)
© 2000 American Society of Agronomy
RICE
Use of Chlorophyll Meter Sufficiency Indices for Nitrogen Management of Irrigated Rice in Asia
F. Hussaina,
K.F. Bronsonb,
Yadvinder-Singhc,
Bijay-Singhc and
S. Pengd
a National Agricultural Research Center, Land Resources, Institute P.O. NIH, Islamabad, Pakistan
b Soil and Water Sciences Div., IRRI, P.O. Box 3127, 1271 Makati City, Philippines, currently Texas A&M University, Texas Agric. Exp. Stn., Route 3, Box 219, Lubbock, TX 79401 USA
c Punjab Agricultural Univ., Dep. of Soils, Ludhiana, India
d Agronomy, Plant Physiology, and Agroecology Div., IRRI, P.O. Box 3127, 1271 Makati City, Philippines
k-bronson{at}tamu.edu
 |
ABSTRACT
|
|---|
Low N fertilizer recovery efficiency (RE) remains a problem in rice (Oryza sativa L.) production in Asia. The chlorophyll meter has been shown to identify when rice is in need of a N top-dressing, that if applied would result in greater agronomic efficiency (AE) of N fertilizer than commonly practiced, pre-set timing schemes. Critical chlorophyll meter readings have been used to match N top-dressing with plant demand for the cultivar IR72, but fixed critical readings may not apply to multiple cultivars and locations. Our objective was to test the approach of using sufficiency indices, calculated from chlorophyll meter readings relative to well-fertilized reference plots, with several rice cultivars in the Philippines and in India by comparing with local fixed N timing recommendations. Chlorophyll meter readings were significantly affected by N management practice, cultivar, and time of application at both locations. Similar rice yields were produced with chlorophyll meter sufficiency indices compared with the fixed N timing treatment with 30 kg less N ha-1 in the wet season in all cultivars at both locations. Compared with fixed-timing, 45 kg less N ha-1 was used with the chlorophyll meter in the Philippines dry season without a yield reduction in three of four cultivars. In conclusion, chlorophyll meter sufficiency indices identified when several rice cultivars at two locations needed a N top-dressing, which resulted in greater AE of N fertilizer than local, pre-set splits.
Abbreviations: AE, agronomic efficiency • IRRI, International Rice Research Institute • RE, recovery efficiency
|
INTRODUCTION
|
|---|
LOW N FERTILIZER recovery efficiency (RE) in flooded rice in Asia has been reported as ranging from 20 to 40% of applied N (De Datta et al., 1987, 1988; Schnier et al., 1990). These estimates, however, were made using 15N-labeled fertilizer, and the simple difference method calculations in some of these same studies indicate that RE by rice ranges from 34 to 64% (Cassman et al., 1993). The low RE of fertilizer-N from Asian flooded rice systems is usually attributed to losses through NH3 volatilization and nitrification–denitrification (Freney et al., 1990).
Recently, however, researchers have demonstrated that in the rice soils of the Philippines with relatively high soil N supply capacity, RE of N fertilizer can be improved by avoiding applications at transplanting when plant demand is low and gaseous loss potential is high (Peng and Cassman, 1998). Recovery efficiency (calculated by the difference method) of urea top-dressed during peak plant demand, such as at panicle initiation, can be as high as 78% (Peng and Cassman, 1998).
The question is whether the critical time for top-dressing can be determined with a rapid, in-field method. Soil tests for N fertilizer recommendations in flooded rice soils have not been shown to be reliable (Stalin et al., 1996; Adhikari et al., 1999). Turner and Jund (1991) demonstrated that the chlorophyll meter, which measures leaf greenness, can predict the need for N top-dressing at prepanicle initiation and panicle differentiation for rice in Texas. The chlorophyll meter has been used successfully by soil scientists in Nebraska for corn by applying in-season N fertilizer doses when sufficiency indices of chlorophyll meter readings of the plot in question divided by a well-fertilized reference plot fall below 0.95 (Peterson et al., 1993; Varvel et al., 1997). The critical sufficiency index of 0.95 corresponded to N deficiency that can lead to a yield reduction in corn if N is not applied. This method has the advantage of being self-calibrating for different soils, seasons, and cultivars.
In the early work on use of the chlorophyll meter in rice production in Asia, Peng et al. (1993) reported that the correlation between chlorophyll meter readings and N concentration of rice leaves was improved by adjusting for specific leaf weight. Peng et al. (1996) then focused on determining a fixed critical chlorophyll meter reading that farmers could refer to in the field. They suggested the use of 35 as a critical chlorophyll meter reading for a recent International Rice Research Institute (IRRI) cultivar, IR72, in the dry season. Top-dressing of 30 kg N ha-1 is recommended when chlorophyll meter readings fall below this number. The 35 reading correlated to 1.4 g N m-2 leaf area, a number that was found to be fairly stable for a high-yielding IR72 crop during the dry growing season. In field trials in the Philippines, use of the 35 critical reading was found to result in higher AE (i.e., similar yields with less N fertilizer applied) compared to fixed split-timing schemes (Peng et al., 1996). IRRI (1995) recommends that rice scientists in each region determine the appropriate critical values for their rice cultivars and seasons (wet or dry). We hypothesized that use of a single chlorophyll meter sufficiency index will result in more efficient use of N fertilizer, without reducing yields, than the currently recommended fixed timing N management plans for different locations, seasons, and cultivars. Our main objective was to test the sufficiency index approach of chlorophyll meter–based N management for rice, based on well-fertilized reference plots for two locations in Asia by comparing with fixed N fertilizer applications.
|
Materials and methods
|
|---|
Field studies were conducted in the 1997 wet season and 1998 dry season on a Guadalupe clay (Aquandic Epiaqualfs) at IRRI, Los Baños, Philippines, and on a Fatehpur loamy sand (Typic Ustipsamments) in the wet season of 1997 in Ludhiana, India. Chemical properties of the two soils are shown in Table 1 . Twenty-day-old seedlings were transplanted in hills on a 20- by 20-cm grid on 23 June and 1 July 1997 in Ludhiana and Los Baños, respectively, for the wet season crops, and on 5 Jan. 1998 for the dry season crop in Los Baños.
View this table:
[in this window]
[in a new window]
|
Table 1 Soil chemical properties of the 0- to 30-cm layer of a Guadalupe clay, Los Baños, Philippines, and a Fatehpur loamy sand, Ludhiana, India
|
|
The experiments were conducted in a randomized complete block design with split-plots. There were four and three replicates in Los Baños and Ludhiana, respectively. The main plot factor was cultivar: IR64, IR72, IR65469, and IR68284H in both seasons at Los Baños, and PR-106 and PR-111 at Ludhiana. The split-plot factor consisted of the following urea-N fertilizer treatments: well-fertilized reference, chlorophyll meter–based N management, fixed timing, and zero N control. Subplots measured 5 by 5 m2.
We decided to use N rates for the well-fertilized plots that were 180 to 200% of the commonly recommended N rate in the fixed timing practice. In the wet and dry seasons of Los Baños, the N fertilizer rates for the well-fertilized reference were 225 and 325 kg N ha-1, respectively. The reason for the high N rates in the well-fertilized plots was to ensure that no N deficiencies occurred in this reference treatment. Our intention was to stimulate luxury uptake of N in these reference plots. High N fertilization does not lead to an excessive production of chlorophyll (Schepers et al., 1992; Peterson et al., 1993). The fact that chlorophyll and tissue N, and therefore chlorophyll meter readings, have an upper limit allows us to use the well-fertilized plot as a reference. Although we anticipated lodging in the well-fertilized plots, this occurs during grain filling, when we were no longer taking chlorophyll meter readings or applying N topdressings.
The fixed timing N fertilizer rates in Los Baños were 120 and 180 kg N ha-1 in the wet and dry seasons, respectively, as recommended by Peng et al. (1996). Higher N fertilizer recommendations for dry season rice compared to wet season rice reflect the higher solar radiation and yield potential of the dry season (Kropff et al., 1993). The total N fertilizer in Los Baños for well-fertilized reference and fixed timing treatments were applied in fourths at final harrowing prior to transplanting, midtillering, panicle initiation, and at early flowering. The fertilizer rates for the well-fertilized and fixed-timing treatments in Ludhiana were 240 and 120 kg N ha-1, respectively, applied in three equal splits at final harrowing, midtillering, and panicle initiation (Bijay-Singh et al., 1991).
Basal N fertilizer applications prior to transplanting at final harrowing in the chlorophyll meter–based treatments were at the same rate as the basal dose in the fixed timing treatment. At final harrowing, all location-years received 22 kg P ha-1 and 83 kg K ha-1 as single superphosphate [Ca(H2PO4)2] and KCl, respectively. Rice seedlings were dipped in 1% ZnO solution for 24 h prior to transplanting (De Datta et al., 1987). Applications of P and K fertilizer were made despite greater than critical soil test values to minimize the chance of development of midseason deficiencies of these elements.
Weekly chlorophyll meter (SPAD-502, Minolta, Ramsey, NJ) readings were taken in all fertilized plots starting 14 d after transplanting. Twenty hills of rice were chosen at random in each plot, and at each hill three readings were taken from the uppermost fully expanded leaf. The sufficiency index was calculated as suggested by Peterson et al. (1993):
 | (1) |
When the sufficiency index fell below 90%, 30 kg urea-N ha-1 was broadcast-applied on the same day. A 90% sufficiency index was chosen instead of the 95% used for corn (Peterson et al., 1993; Varvel et al., 1997) based on earlier work by Peng et al. (1996). They reported chlorophyll meter readings in their chlorophyll meter–based treatment that were often less than 90% of the readings of the fixed-timing treatments that received more N. Additionally, we considered that at comparable leaf N concentrations (e.g., 35 g N kg-1), corresponding chlorophyll meter readings are about 55 and 35 in corn and rice, respectively (Schepers et al., 1992; Peng et al., 1993). Therefore, in the less green leaves of rice, resolving readings in the chlorophyll meter–based N management treatment from the well-fertilized reference plot readings would be easier with a 90% sufficiency index than with a 95% index. Chlorophyll readings were taken up to the time of 50% flowering.
At physiological maturity, plants growing in an area of 5.8 m2 (145 hills) were cut at ground level by hand and mechanically threshed. Moisture content of rice grain was recorded, and reported weights were adjusted to 140 g H2O kg-1. Straw and dried leaf weights were randomly taken on 12 of the 145 sampled hills. Nitrogen content of grain and straw was determined by dry combustion (C/N Analyzer, Europa Scientific, Crew, UK).
The N use efficiency measures, RE and AE, were calculated as follows. Recovery efficiency (RE) was calculated as (Dilz, 1988):
 | (2) |
where TNU is the total N uptake in grain and straw. Agronomic efficiency (AE) was calculated as (Novoa and Loomis, 1981):
 | (3) |
Statistical analysis of data consisted of analysis of variance of rice yield parameters for a split-plot design for each location-year. Chlorophyll meter readings were analyzed for each location-year in a repeated measures analysis of variance with week of sampling in the split-split plot (SAS Inst., 1996). Duncan's Multiple Range Test was used at the 0.05 level of probability to test differences between treatment means.
|
Results and discussion
|
|---|
Wet Season 1997, Los Baños, Philippines
Grain yield and total N uptake responded positively to N fertilizer in all treatments and cultivars relative to the control (Table 2) . The main effect of cultivar and the cultivar x N treatment interaction were not significant for yield, N uptake, RE, or AE; therefore, these data are shown averaged across cultivars. A total of 90 kg N ha-1 was applied in the chlorophyll meter–based treatment, 30 kg N ha-1 less than in the fixed timing treatment, but grain yields were not different. Agronomic efficiency was higher in the chlorophyll meter–based treatment compared to the fixed-timing treatment. Typically, AE is greater when less N fertilizer is used, but it is significant that this objective was achieved in the chlorophyll meter–based treatment without reducing yields. The RE of about 60% and AE of 18 to 23 kg grain kg-1 N applied for these two treatments are similar to those reported by Peng et al. (1996) for rice grown in Los Baños. In the well-fertilized plots, a mild infestation of sheath blight (Rhizoctonia solani) was observed and lodging was noticed during grain filling. It did not appear that the sheath blight affected chlorophyll meter readings, and the lodging occurred after chlorophyll meter readings were terminated.
View this table:
[in this window]
[in a new window]
|
Table 2 Rice grain yield, total N uptake, recovery efficiency, and agronomic efficiency as affected by N management (averaged across cultivar) in Los Baños, Philippines, during the wet season in 1997
|
|
Analysis of variance of the chlorophyll meter readings indicated significant cultivar, N management, cultivar x N management, week, cultivar x week, and N management x week effects (data not shown). The N management x week interaction is shown in Fig. 1
. Chlorophyll meter readings of the well-fertilized reference treatment, chlorophyll meter–based N management, fixed N timing treatment, and the 90% sufficiency index (chlorophyll meter readings that correspond to a 90% sufficiency index) are compared. At 38 and 59 d after transplanting, readings below the 90% sufficiency index readings were observed in the chlorophyll meter–based N treatments in all cultivars, at which times 30 kg N ha-1 topdressings were made. The amounts that the chlorophyll meter readings fell below the 90% sufficiency index readings varied from less than 0.1 units for IR72 at 38 d to 3.7 units for IR65469 at 59 d. The readings in the chlorophyll meter–based management treatments, however, were statistically lower than the well-fertilized plot readings on these two days. Chlorophyll meter readings with the fixed N fertilizer timing were greater than the 90% sufficiency index readings throughout the season, suggesting sufficient N. Chlorophyll meter readings for each cultivar corresponding to the 90% sufficiency index decreased in the order IR72 > IR64 = IR68284H > IR65469. This confirms our hypothesis that N management strategies need to accommodate cultivar differences, in contrast to the present blanket N recommendation in the region. The chlorophyll meter reading corresponding to the 90% sufficiency index was greater than the critical level of 35 suggested by Peng et al. (1996) for most of the season, indicating that sufficiency indices may be more suitable across seasons than fixed critical readings.
View larger version (22K):
[in this window]
[in a new window]
|
Fig. 1 Chlorophyll meter readings as affected by cultivar and N management in Los Baños, Philippines, during the wet season in 1997. The 90% sufficiency index readings were calculated as 90% of the well-fertilized plot readings. Arrows indicate time of 30 kg N ha-1 top-dressing to chlorophyll meter–based treatments
|
|
Wet Season 1997, Ludhiana, India
Grain yields and total N uptake (uptake data not shown) in the N-fertilized treatments were higher than in the zero-N control (Table 3)
. Varietal differences in yield were observed, but not in N uptake, and there were no cultivar x N interactions. Consistent with the results from the wet season study in the Philippines, although 90 and 120 kg N ha-1 were applied to both cultivars in the chlorophyll meter–based and fixed-timing treatments, respectively, grain yields were similar. Urea-N topdressings of 30 kg N ha-1 were applied at 21, 28, 57, and 65 d after transplanting in PR-106 and PR-111, when readings in the chlorophyll meter–based treatment fell below the 90% sufficiency index (data not shown). Agronomic efficiency, but not RE of N, was improved by chlorophyll meter–based N management compared with fixed timing, again without reducing yields (data not shown).
View this table:
[in this window]
[in a new window]
|
Table 3 Rice grain yield as affected by N management and cultivar in Ludhiana, India, during the wet season in 1997
|
|
Dry Season 1998, Los Baños, Philippines
Grain yields were greater in all N-fertilized treatments compared with the zero-N control, and a N x cultivar interaction was observed (Table 4)
. In all four cultivars, 135 kg N ha-1 was applied in the chlorophyll meter–based treatment, 45 kg N ha-1 less than in the fixed-timing treatment. Except for IR64, grain yields were similar in the chlorophyll meter–based and the fixed-timing treatments. In IR64, grain yield and total N uptake were depressed with chlorophyll meter–based management compared with fixed timing. Correspondingly, AE was greater with chlorophyll meter management in all cultivars except IR64 (data not shown). Agronomic efficiency in this study was slightly lower than reported by Peng et al. (1996) in the dry season at this location, reflecting the lower yields in this study. A greater number of cloudy days and lower temperatures than normal for the 1998 Philippine dry season can explain the low dry season yields.
View this table:
[in this window]
[in a new window]
|
Table 4 Rice grain yield as affected by N management and cultivar in Los Baños, Philippines, during the dry season in 1998
|
|
|
Conclusions
|
|---|
These preliminary data indicate that a well-fertilized reference plot and a chlorophyll meter may be used as an indicator of in-season N status of irrigated rice in Asia. Our hypothesis, that the sufficiency index approach adapts to different seasons, soil types, and cultivars, was validated. Our objective of producing similar grain yields to the fixed-timing treatment with the chlorophyll meter using less N fertilizer was achieved in 9 of 10 location–cultivar–season combinations tested. The choice of N rates for the well-fertilized reference plots of 1.8 to 2 times the recommended rates was reasonable, since grain yields were generally similar to those obtained with recommended rates but N uptake was greater.
The relatively high price of the instrument remains an issue regarding the use of the chlorophyll meter in Asian rice farming. The Crop Resource Management Network at IRRI is facilitating on-farm testing of a Leaf Color Chart, which consists of six color chips of a range of greenness (IRRI, 1996). While the leaf color chart is not as precise or as accurate as the chlorophyll meter, it does have an obvious cost advantage. The approaches we have established in rice with chlorophyll meter sufficiency indices and the fixed critical readings of Peng et al. (1996) are an important foundation of methodology upon which future testing and development in research stations and in farmers' fields can build.IRRI 1995; SAS Institute 1996
|
NOTES
|
|---|
Contribution of the International Rice Research Institute and Punjab Agricultural University.
Received for publication March 13, 1999.
|
REFERENCES
|
|---|
- Adhikari C., Bronson K.F., Panuallah G.M., Regmi A.P., Saha P.K., Dobermann A., Olk D.C., Hobbs P.R., Pasuquin E. On-farm soil N supply and N nutrition in the rice–wheat system of Nepal and Bangladesh. Field Crops Res. 1999;64:273-286.
- Bijay-Singh Yadvinder-Singh, Khind C.S., Meelu O.P. Leaching losses of urea-N applied to permeable soils under lowland rice. Fert. Res. 1991;28:179-184.
- Cassman K.G., Kropff M.J., Gaunt J., Peng S. Nitrogen use efficiency of rice reconsidered: What are the key constraints?. Plant Soil 1993;155:359-362.
- De Datta S.K., Buresh R.J., Samson M.I., Kai-Rong W. Nitrogen use efficiency and nitrogen-15 balances in broadcast-seeded flooded and transplanted rice. Soil Sci. Soc. Am. J. 1988;52:849-855.[Abstract/Free Full Text]
- De Datta S.K., Obcemea W.N., Chen R.Y., Calabio J.C., Evangelista R.C. Effect of water depth on nitrogen use efficiency and nitrogen-15 balance in lowland rice. Agron. J. 1987;79:210-216.[Abstract/Free Full Text]
- Dilz K. Efficiency of uptake and utilization of fertilizer nitrogen by plants. In: Jenkinson D.S., Smith K.A., eds. Nitrogen efficiency in agricultural soils. London: Elsevier Applied Science, 1988:1-26.
- Freney J.R., Trevitt A.C.F., De Datta S.K., Obcemea W.N., Real J.G. The interdependence of ammonia volatilization and denitrification as nitrogen loss processes in flooded rice fields in the Philippines. Biol. Fertil. Soils 1990;9:31-36.
- IRRI. Use of chlorophyll meter for efficient N management in rice. Manila, Philippines: IRRI, 1995 Crop Resource Management Network Technology Brief No. 1..
- IRRI. Use of Leaf Color Chart (LCC) for N management in rice. Manila, Philippines: IRRI, 1996 Crop Resource Management Network Technology Brief No. 1..
- Kropff M.J., Cassman K.G., van Laar H.H., Peng S. Nitrogen and yield potential of irrigated rice. Plant Soil 1993;155:391-394.
- Novoa R., Loomis R.S. Nitrogen and plant production. Plant Soil 1981;58:177-204.
- Peng S., Cassman K.G. Upper thresholds of nitrogen uptake rates and associated nitrogen fertilizer efficiencies in irrigated rice. Agron. J. 1998;90:178-185.[Abstract/Free Full Text]
- Peng S., Garcia F.V., Laza R.C., Cassman K.G. Adjustment for specific leaf weight improves chlorophyll meter's estimate of rice leaf nitrogen concentration. Agron. J. 1993;85:987-990.[Abstract/Free Full Text]
- Peng S., Garcia F.V., Laza R.C., Sanico A.L., Visperas R.M., Cassman K.G. Increased nitrogen use efficiency using a chlorophyll meter in high-yielding irrigated rice. Field Crops Res. 1996;47:243-252.
- Peterson, T.A., T.M. Blackmer, D.D. Francis, and J.S. Schepers. 1993. Using a chlorophyll meter to improve N management. Nebguide G93-1171A. Coop. Ext. Serv., Univ. of Nebraska, Lincoln.
- SAS Institute. SAS system for Windows. Cary, NC: SAS Inst, 1996 Release 6.12..
- Schepers J.S., Francis D.D., Virgil M., Below F.E. Comparison of corn leaf nitrogen concentration and chlorophyll meter readings. Commun. Soil Sci. Plant Anal. 1992;23:2173-2187.
- Schnier H.F., Dingkuhn M., De Datta S.K., Marquesses E.P., Faronilo J.E. Nitrogen-15 balance in transplanted and direct-seeded flooded rice as affected by different methods of urea application. Biol. Fertil. Soils 1990;10:89-96.
- Stalin P., Dobermann A., Cassman K.G., Thiyagarajan T.M., Ten Berge H.F.M. Nitrogen supplying capacity of lowland rice soil of southern India. Commun. Soil Sci. Plant Anal. 1996;27:2851-2874.
- Turner F.T., Jund M.F. Chlorophyll meter to predict nitrogen topdress requirement for semidwarf rice. Agron. J. 1991;83:926-928.[Abstract/Free Full Text]
- Varvel G.E., Schepers J.S., Francis D.D. Ability for in-season correction of nitrogen deficiency in corn using chlorophyll meters. Soil Sci. Soc. Am. J. 1997;61:1233-1239.[Abstract/Free Full Text]
This article has been cited by other articles:
|
|
|
|
 
S. M. Samborski, N. Tremblay, and E. Fallon
Strategies to Make Use of Plant Sensors-Based Diagnostic Information for Nitrogen Recommendations
Agron. J.,
July 7, 2009;
101(4):
800 - 816.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
G. Stevens, A. Wrather, M. Rhine, E. Vories, and D. Dunn
Predicting Rice Yield Response to Midseason Nitrogen with Plant Area Measurements
Agron. J.,
February 26, 2008;
100(2):
387 - 392.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
K. F. Bronson, J. D. Booker, J. W. Keeling, R. K. Boman, T. A. Wheeler, R. J. Lascano, and R. L. Nichols
Cotton Canopy Reflectance at Landscape Scale as Affected by Nitrogen Fertilization
Agron. J.,
April 27, 2005;
97(3):
654 - 660.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
T. T. Chua, K. F. Bronson, J. D. Booker, J. W. Keeling, A. R. Mosier, J. P. Bordovsky, R. J. Lascano, C. J. Green, and E. Segarra
In-Season Nitrogen Status Sensing in Irrigated Cotton: I. Yields and Nitrogen-15 Recovery
Soil Sci. Soc. Am. J.,
September 1, 2003;
67(5):
1428 - 1438.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
K. F. Bronson, T. T. Chua, J. D. Booker, J. W. Keeling, and R. J. Lascano
In-Season Nitrogen Status Sensing in Irrigated Cotton: II. Leaf Nitrogen and Biomass
Soil Sci. Soc. Am. J.,
September 1, 2003;
67(5):
1439 - 1448.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
B. Singh, Y. Singh, J. K. Ladha, K. F. Bronson, V. Balasubramanian, J. Singh, and C. S. Khind
Chlorophyll Meter- and Leaf Color Chart-Based Nitrogen Management for Rice and Wheat in Northwestern India
Agron. J.,
July 1, 2002;
94(4):
821 - 829.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
K. F. Bronson, A. B. Onken, J. W. Keeling, J. D. Booker, and H. A. Torbert
Nitrogen Response in Cotton as Affected by Tillage System and Irrigation Level
Soil Sci. Soc. Am. J.,
July 1, 2001;
65(4):
1153 - 1163.
[Abstract]
[Full Text]
[PDF]
|
|
|
TOP
ABSTRACT
INTRODUCTION
