Introduction

The production of milk in the district Lagunera is intensive and highly tecnificada producing around 9 million daily liters (^{SAGARPA, 2015}) and one of the principal raw materials for the cattle's nutrition it is maíz's silage. The alfalfa and corn are the principal fodder cultivations at this geographic zone, they complement both cultivations in relation to the nutrition of the animals, the corn in the contents of fiber required by the ruminants for digestion and the alfalfa like source of protein in the production of milk (^{Montemayor et al., 2012}). The average yields of green meat and I dry of corn in the district Lagunera they come from 49 and 17 Mg ha^-1 is respectively (^{Cueto et al., 2006}).

One of the main limiting factors for obtaining high yield and quality of corn silage is the dose of nitrogen fertilization (^{de Menezes et al., 2013}). Because of this it is common for producers applied dose of nitrogen (N) higher than those required by the culture. This increases production costs, reduces crop profitability and increasing potential risk of contamination as the N does not absorb the plant is lost through different processes, such as volatilization, denitrification and nitrate leaching etc. (^{Delgado y Follett, 2010}). To reduce the risk of nitrate pollution significantly without reducing yields is necessary to know the nutrient requirements of crops, fertilizer type, shape and phenological stage this input should be applied (^{Klocke et al., 1999}).

The efficient management of nitrogen fertilizers in agricultural production systems is essential to avoid excessive applications, avoiding potential risks of environmental pollution without reducing the profitability of the crop (^{Fallah y Tadayyon, 2010}; ^{Cueto et al., 2013}). Because the importance of N in plants, the high amounts being sued in the growth of crops and the growing concern for preserving the environment, was raised as targets for this work. 1) evaluate the efficiency with which the forage maize used to produce the N dry matter; and 2) determine whether the efficient use of nitrogen (UEN) and the apparent recovery of nitrogen (RAN) are affected by soil type and dose of N.

Materials and methods

The study was conducted in the Experimental La Laguna (CELALA) of the National Institute of Forestry, Agriculture and Livestock (INIFAP) in Matamoros, Coahuila, located in the Laguna Region. This region is geographically located at latitude 24° 22' north latitude and the meridian 102° 22' W, has an altitude of 1 120 meters, the annual average rainfall is 215 mm and the average annual temperature 22.6 °C (^{INEGI, 2010}). two soils of contrasting texture were used: clayey and sandy texture up to 60 cm deep; in Table 1, some other features of these soils is. All soil analyzes were performed according to the methods described in the Official Mexican Standard NOM-021-RECNAT-2000 (^{SEMARNAT, 2000}). Forage maize variety SB-302 was used, which was sown in plots of 10 m wide by 15 m long.

The experimental design was randomized complete block with four replications. All plots received 80 kg ha^-1 of P₂O₅, as a source phosphoric acid (52% P₂O₅). N doses evaluated are noted in Table 2. The dose of 300 kg ha^-1 was calculated using a balance method (^{Palma et al., 2002}) for a performance target of 16 Mg ha^-1 MS, considering extraction crop of 14 kg N Mg MS (^{Faz et al., 2006}), residual N in soil 75 kg ha^-1 and N use efficiency of 60%.

Other treatments accounted for 0, 33, 67 and 133% of the dose of N + N residual (0, 50, 175 and 425 kg ha^-1), doses were divided into three applications before each of the first three irrigations (^{Cueto et al., 2013}). The percentage of the dose of N applied in each irrigation is noted in Table 3. The source was ammonium N (20.5% N), which was dissolved in water before applying irrigation sulfate. A system of piping with multi-floodgates to apply the irrigation was utilized.

Was applicable to him an irrigation of pre-planting or aniego, and four irrigations of help, to the 21 40 57 and 73 days after planting (DDS). The harvest came true to the 97 DDS, in the maturity stage of the grain of a third part of advance of the milk line (Nuñez et al., 2005). The forage yield was estimated to be a useful plot two center rows 10 m long. To express the dry matter yield, the moisture content was calculated on a sample of four plants per plot, which were weighed fresh and then oven dried to constant weight; MS average was 37.5%. For analysis of N plan a sample of the aerial part of five plants per plot was taken and the total N was analyzed in the whole plant by the Kjeldahl method (^{Jones, 2001}). The variables analyzed were: total yield of dry matter (MS), extraction and concentration of N in the plant efficiency and apparent recovery N.

Efficient use of N (SBU) is calculated with Equation 1: (^{Zemenchik and Albrecht, 2002}).

Where: UEN= efficient use of nitrogen (ML Mg kg^-1 N), MSTi= dry matter harvested in the fertilized treatment (Mg ha^-1), MST₀= dry matter harvested in the treatment without nitrogen (Mg ha^-1) and DN= dose nitrogen fertilized treatment (kg ha^-1).

The apparent recovery of N (RAN) was calculated using Equation 2: (^{Zemenchik and Albrecht, 2002}).

Where: RAN= apparent recovery of nitrogen (%), Net_i= nitrogen extracted by the plant in the fertilized treatment (kg ha^-1), NET₀= nitrogen extracted by the plant in the treatment without nitrogen (kg ha^-1) and DN= dose of nitrogen fertilization treatment (kg ha^-1).

The data obtained were analyzed with the statistical program SAS version 9.1 (^{SAS Institute, 2003}), by analysis of variance and comparison of treatment means was performed using Duncan's test with a significance level of 0.05.

Results and discussion

Nitrogen extracted

The extraction of N in the two soils was significantly lower in the treatment where no N was applied with respect to the fertilized treatments (Table 5).

Table 5 Nitrogen extracted by the plant (kg ha^-1). 

Tratamientos	Arcilloso	Arenoso
0	35.25 c†	83.36b
50	216.9b	161.65a
175	238.02 ab	134.14a
300	217 15b	186 14a
425	267.85 a	183.57a

^†Letras distintas en la misma columna indican diferencias significativas (Duncan, 0.05).

The highest average extraction of N was in the clayey soil. The increase in the dose of nitrogen in the clay soil of 50 to 425 kg ha^-1 caused significant increases in the extraction of N, with no differences for sandy soil as the latter is reduced fertility and increased susceptibility to suffer leaching losses because the ammonium in clay soil can be retained in the exchange sites which partially prevents loss (^{Salgado et al., 2001}). In our study the lowest dose of fertilization, extraction exceeded the applied dose, indicating that a large portion of plant N was provided on the floor; in constrast with increasing fertilization, extraction decreases indicating that the N is not properly utilized by the crop, thus reducing the UEN (Figure 1). The increase in the extraction of N is related to greater production of MS (Table 3). The values in the extraction of N are similar to those found by other studies in maize ^{Menezes et al. (2013)} report a removal of N (129-224 kg ha^-1) with a dry matter yield of 14.8 to 19.9 Mg ha^-1; instead ^{Butler et al. (2008)} reports a removal of N (176- 311 kg ha^-1) with MS yields from 13 to 25 Mg ha^-1.

Figure 1 Efficient use of nitrogen (UEN) kg MS/kg N in different N fertilization rates in 2007 and 2008 for the two types of sandy and clay soil. 

The N extracting the plant at harvest, depending on the dose of N to a quadratic model, significant r² values of 0.75 to 0.84 was adjusted (Table 6).

Table 6 Regression equations and coefficient of determination of nitrogen removal at harvest (N_ext) depending on the dose of nitrogen. 

Suelo	Ecuación	r2
Arcilloso	Next=-0.0196•N2+4.0004•RN + 58.914	0.8379
Arenoso	Next=-0.005•N2+ 1.371•RN +93.34	0.758

Apparent recovery of N

The RAN represents the percentage of nitrogen applied as fertilizer, which is extracted by aerial or crop harvested and used as an indicator of the availability of N (^{Zemenchik and Albrecht, 2002}). The fertilization rate of 50 kg ha^-1, in both obtuvienon a RAN, statistically higher than the other treatments, most RAN was in clay soil with 363% soil while in sandy soil RAN was 157 (Figure 2), however at higher levels of fertilization decreases as the RAN requirements crop N exceed (^{Cogger et al., 2001}).

Figure 2 Apparent recovery of nitrogen in different doses of N fertilization on clay and sandy. 

The determination of these indices can help reduce application rates of nitrogen fertilization carried out in the cultivation of forage maize. RAN values, depending on the dose of N is adjusted to a quadratic model, with coefficients of determination of from 0.82 to 0.97.

Conclusions

Efficient use of nitrogen and the apparent nitrogen recovery are affected by soil texture and fertilization. The highest values in the efficient use of nitrogen and the apparent nitrogen recovery were higher in the clayey soil. With the lowest dose of fertilization (50 kg ha^-1), the largest amount of N recovered, with 29 kg MS kg N^-1 in clay soil and 11 kg MS kg N^-1 on the sandy, the apparent recovery It exceeds 56% nitrogen in the soil of loamy sandy texture to texture. The values in the efficient use of nitrogen and the apparent nitrogen recovery, decrease with increasing fertilization. To prevent nitrogen loss without reducing biomass production in forage maize under the conditions of this study we recommend avoiding higher doses at 50 kg N ha^-1.