NITROGEN FERTILIZATION AND SOIL MANAGEMENT OF WINTER COMMON BEAN CROP

The adoption of appropriate cultural management, which includes nitrogen fertilization and soil tillage system, is very important to increase the efficiency of plant in the utilization of available resource. This research work was conducted for three years aiming at evaluating the effect of sidedressing nitrogen application (0, 25, 50, 75, and 100 kg ha) on winter bean crop under different systems of soil management (conventional, minimal, and no-tillage systems). The experimental design was a randomized block arranged in strips relative to tillage systems, with random distribution of nitrogen doses into each strip, with four replications. Grain yield was affected by nitrogen rates and significant increases were obtained with the application of 75 to 100 kg N ha. Soil management did not affect grain yield, although the “minimum system” provided better results in the two first years.


INTRODUCTION
Although the common bean plants have N-fixation ability by symbiosis with Rhizobium, they are recommended to be nitrogen fertilized, because N-fixation does not provide enough N to supply enough N to meet requirement of this crop (Coelho et al., 1998).The recommended N rate for common beans in the Cerrado (savanna) region of the States of São Paulo, Minas Gerais, Paraná, Espírito Santo and Santa Catarina, varies from 30 to 60 kg ha -1 , based on grain yield response curves for N, since there is no correlation between grain yield and soil-organic matter or soil-N (Moraes, 1988).
Maximal grain yield was obtained for irrigated common beans cultivated in the region of Ilha Solteira, with the application of 63 kg ha -1 of sidedressed N (Arf et al., 1992).On the other hand, increases in grain yield and other bean yield components such as total dry matter, shoot-N concentrations and contents, were obtained during winter season, for N-fertilization rates up to 90 kg ha -1 , in a field experiment in the State of Goiás, in a Typical Acrustox (Silveira & Damasceno, 1993).The use of higher rates of N has provided higher bean grain yields (Araújo et al., 1994;Barbosa Filho & Silva, 2000;Guerra et al., 2000).The application of 75 kg ha -1 of N has resulted in 38% increase in productivity, applied either in the sowing row or sidedressed, and increased total winter crop yield in the West Region of State of São Paulo (Carvalho et al., 2001).Soil tillage is a management practice as important as fertilization in what concerns the conservation of soil chemical, physical and biological properties.Castro et al. (1986), studying soil tillage systems at Pindorama and Campinas, municipal districts of the State of São Paulo, concluded that soil tillage with moldboard-plow resulted in superior grain yields when compared to the conventional tillage systems (leveling-disk and diskplow).Even though traditional tillage systems have been considered adequate for the common bean crop it is necessary to clearly define which of the traditional methods is the most interesting and efficient one (Kluthcouski et al., 1988;Sampaio et al., 1989;Siqueira et al., 1993).Some of the soil tillage systems may not affect common bean yielding (Zaffaroni et al., 1991;Landers, 1995;Stone & Moreira, 2000).However, the no-tillage system was reported to be superior to the conventional systems in some situations, resulting in higher productivities (Silva et al., 1996;Silveira et al., 1999;Stone & Silveira, 1999).Despite the relevance of such results, the use of the disk harrow may provide higher productivities for winter-irrigated common beans cultivated at the Central Region of the State of Goiás (Silveira et al., 2001) in comparison to other tillage systems.The authors reported, in addition, a decrease in bean yielding when the no-tillage system was continuously used.The objective of this research was thus evaluating the effect of sidedressed nitrogen fertilization and soil tillage practices on winter irrigated common bean yielding.

MATERIAL AND METHODS
The experiment was carried out at the municipal district of Selvíria, State of Mato Grosso do Sul, Brazil (51 o 22' W, 20 o 22' S; altitude 335 m), during the winter period of 2000, 2001 and 2002.The soil was a Typic Acrustox (USDA, 1996).Average annual rainfall of 1370 mm, average annual temperature of 23.5 o C and average annual relative humidity of 70-80% characterizes the region climate.The soil chemical attributes were determined in soil samples from the 0-0.20 m layer before the experiment, according to Raij & Quaggio (1983), and results were as follows: pH (0.01 mol L -1 CaCl 2 ) = 5.7; O.M.= 22 g cm -3 ; K, Ca, Mg, H+Al and Al = 1.1; 31.0;9.0; 20.0 and 0.0 mmol c dm -3 , respectively; and V = 67%.The experimental design was a randomized block arranged in soil strips and the treatments consisted of five N-rates and three soil-tillage systems.The sidedressing N-rates applied were: 0, 25, 50, 75, and 100 kg ha -1 .The soil tillage systems consisted of (a) disk-plow + disk harrow = conventional; (b) moldboard-plow + disk harrow = minimal tillage; and (c) no-tillage.This latter system was studied in a 5-year-no-tillage area previously cultivated with maize, Tractor cultivar, with 0.90 m row spacing.The maize crop sowing was in January, after adequate fertilization rate recommended by Raij & Cantarella (1996).After maize harvesting, the no-tillage area was prepared for the common bean cropping.A basic fertilizer rate was applied in the sowing rows according to the soil analysis and recommendations of Ambrosano et al. (1996), which consisted of 250 kg ha -1 of a 08-28-16 (N-P 2 O 5 -K 2 O) formula.The N-rates were sidedressed in a unique application, using urea, three weeks after seedling emergence.The plots consisted of six, 6 m-rows, experimental units were the four central, 5 m-rows.Experiments used the common bean 'IAC-Carioca-Eté' cultivar, sowed in the third week of May, using sowing machine adjusted to 12-13 plants m -1 and 0.50 m row spacing (IAC, 1999).Weeds, mainly Bidens pilosa, Commelina benghalensis, Panicum maximum Jacq., Ipomea grandifolia (Dammer) O'Don and other gramineae, were controlled using fluazifop-p-butil + fomesafen herbicide mixture (200 + 250 g a.i.ha -1 ).Pests, mainly Bemisia tabaci and Bemisia argentifolii, Diabrotica speciosa and other coleoptera, were controlled using the methamidophos (300g a.i.ha -1 ) and thiametoxan (250g a.i.ha -1 ).It was also applied mancozeb (1600g a.i.ha -1 ), benomyl (250g a.i.ha -1 ) and oxycarboxin (375g a.i.ha -1 ).Field irrigation was done using conventional, water-spraying system.
The following variables were evaluated: (1) number of days from seedling emergence to full-flowering (50% of experimental unit plant flowering); (2) shoot dry matter yields at full-flowering: 10 plants were collected from the experimental unit, dried in a forced-air oven at 60-70 o C until constant mass and weighed; (3) shoot nitrogen concentration at full-flowering: 10 plants were collected from the experimental unit, dried in a forced-air oven at 60-70 o C during 72h, ground in a Willey-type grinder, followed N determination in plant samples, after sulfuric digestion according to Sarruge & Hagg (1974); (4) yield components at harvest: 10 plants were collected from the experimental unit and evaluated for the average pod number per plant; average grain number per pod, and average grain number per plant; (5) average 100grain mass (from 2 random samples of 100 grain per experimental unit); (6) cycle: number of days from seedling emergence to harvesting, evaluated in all plants from two 5 m-rows of the experimental unit; and (7) these latter plants were submitted to mechanical cleansing to determine grain yield expressed in kg ha -1 (with 13% moisture).Data were analyzed separately for each cropping year, since the variances among years were out of the limit variation recommended for satisfactory grouping analysis (Pimentel Gomes, 1985).

RESULTS AND DISCUSSION
During the three years of field tests, the common bean seedling emergence was uniform, with 13 plants m -1 , as expected.Also, for the three experiments, the full flow- The soil management treatments did not affect plant dry matter yields (Table 1) in any of the three years.These results do not confirm the tendency for higher dry matter yield values previously observed for the no-tillage system (Urchei et al., 2000), but agree with results obtained by Stone & Moreira (2001), who did not observe any effect of soil tillage on this variable.Concerning effects of nitrogen rates on plant dry matter yields, a significant increase was observed only during the first year, and data fitted a quadratic regression equation (Figure 1).During the two following years, the application of 75 to 100 kg ha -1 of N resulted in the highest dry matter yields, corroborating similar positive results obtained by Silveira & Damasceno (1993), Lima et al. (2001), and Stone & Moreira (2001).Nevertheless, sometimes the dry matter yields of common beans do not increase in response to application of sidedressed N-rates (Arf et al., 1994).
Higher shoot N concentration values were observed in plants from the no-tillage treatment in the first year (Table 1), and were within the adequate limit range for common beans according to the literature.Similar results, that is, higher plant N concentration in the no-till-age system, were also obtained by Silveira & Damasceno (1993).However, the tillage system may not always affect this variable (Stone & Moreira, 2001).The N rates resulted in increasing shoot N concentration in the years 2001 (linear effect) and 2002 (quadratic effect) (Figure 2), what might have resulted from higher available N concentration in the soil.Maximum shoot N concentration obtained was 39.9 g kg -1 , and corresponded to soil N application of 68.4 kg ha -1 .Stone & Moreira (2001) also observed increasing shoot N concentrations with increasing N rates applied to the soil.
Number of pods and grains per plant were not affected by the tillage treatments (Table 2).Such results disagree with those obtained by Stone & Moreira (2001), who observed lower pod number per plant in the no-tillage treatment.On the other hand, experiments on soil management using several methods (no-tillage, moldboard-plow, and disk harrow) have shown higher pod number per plant in the no-tillage treatment in comparison to the other systems, allowing infer to that the pod number may vary not only with the soil tillage system but also with other yield component factors, such as cropping year and related environmental conditions (Stone & Moreira, 2000), or else, with the cropping microclimate itself.There were consecutive decreases in both variables along the three cropping years, for all tillage systems used (Table 2).Means followed by common letters in the columns do not differ by Tukey test (0.05).Nitrogen rates increased linearly number of pods seeds number per plant in the years 2000 and2002 (Figures 3 and4), evidencing a direct effect of N-fertilization on these variables.
The soil tillage systems did not affect the number of grains per pod in the years 2000 and 2002.Only during 2001, the use of moldboard-plow + leveling disk (Table 3) resulted in higher grain number per pod.These results disagree with those obtained by Stone & Moreira (2000), who concluded that the no-tillage system provided the highest number of grains per pod, in comparison to the conventional methods.In relation to the N application, a linear increase in the number of grains per pod with the N rates was observed only during the year 2001 (Figure 5).
The 100-grain mass was higher only in 2001 for the minimal tillage treatment (Table 3).This might be explained by the lower number of pods per plant which occurred in the same year and treatment.The no-tillage was the soil management treatment that provided the lowest values for this variable.As a matter of fact Stone & Moreira (2001) reported that the no-tillage system might occasionally result in lower 100-grain mass values.In 2001 and 2002, linear increases in the 100-grain mass values were obtained with increasing N rates (Figure 6), corroborating results previously observed by Stone & Moreira (2000) and Arf et al. (1994).
Variation in grain yield with the soil tillage system was only observed during 2001 (Table 4), when the --------------------g --------------------  highest grain yields were observed for the minimal tillage, in agreement with the results reported by Castro et al. (1986), who observed the superiority of the moldboard-plow method of soil tillage in comparison to the conventional methods for common bean yielding.However, during 2000 and 2002, no differences among the soil tillage systems were observed for the bean grain yield, which averaged 2,000 kg ha -1 .Nevertheless, it is important to point out that the no-tillage system may provide higher bean grain yields than the conventional systems in areas where the presence of straw on soil surface is important to maintain adequate soil moisture and minimize soil temperature fluctuations, favoring retention of N in the soil (Silveira et al., 1999;Carvalho et al., 1999;Stone & Moreira, 2000).This was not observed in the present experiment, since rainy summer and high temperatures characterize the region.This condition accelerates straw decomposition and exposes the soil to weathering.
Despite the linear effect of N rates on grain yield being only noticed in the second year (Figure 7), the application of 75 to 100 kg N ha -1 provided considerable increases in grain yield in the three years.In the second year, the higher number of grains per pod and higher 100seed mass values obtained in response to N-rates were probably the yield components responsible for the increased productivity.
occurred approximately at 40 and 90 days after seedling emergence, respectively.No significant interactions were observed among the evaluated variables.

Figure 1 -
Figure 1 -Variation of shoot dry matter with increasing nitrogen rates.2000

Figure 4 -Figure 3 -
Figure 4 -Variation of the number of grains per plant with increasing nitrogen rates.

Figure 5 -
Figure 5 -Variation of the number of grains per pod with increasing nitrogen rates.2001.

Figure 6 -
Figure 6 -Variation of mass of 100 grains with increasing nitrogen rates.

Figure 7 -
Figure 7 -Grain yield increase with the increasing nitrogen rates, 2001.

Table 1 -
Shoot dry matter yields and N-concentrations at flowering of a common bean crop under different soil tillage systems.

Table 2 -
Number of pods and grains per plant of a common bean crop under different soil tillage systems.

Table 3 -
Number of grains per pod and mass of 100 grains of a common bean crop under different soil tillage systems.
Means followed by common letters in the columns do not differ by Tukey test (0.05).