Introduction

The Highland valleys physiography with acid soils of volcanic neo chain between 2 200-2 700 msnm and is where more than one million hectares of maize grown in the states of Tlaxcala, Puebla, Hidalgo, Querétaro, Michoacán and the State from Mexico. In the latter state corn is grown of march to October rains 600-800 mm. Yields averaged in this state per hectare of 2.65 t ha^-1 in temporary and 3.75 t ha^-1 with irrigation (^{Trueba, 2012}). In recent years, grain yields of corn in Highland valleys State of Mexico have stalled, at which strategies are explored to improve the situation, including some alternatives, highlighting reports on genetic improvement and adoption of best practices fertilization and management in maize in Highland valleys.

To increase grain yields of maize has gone to the reciprocal recurrent selection in maize populations Highland valleys in soils with high and low nitrogen content (^{MorenoPérez et al., 2004}), evaluation of the effect of the level of moisture and soil nitrogen in the behavior of hybrid corns and native (^{Kibet et al., 2009}); the study of normal corn hybrids and high quality protein to chemical fertilization (^{Palafox et al., 2005}); and planting of improved seeds and nutrition, primarily with nitrogen (^{Barbieri et al., 2008};. and ^{Cervantes et al., 2013}).

For the efficiency of utilization of nutrients two strategies are followed: one to increase the absorption and assimilation of N by the plant for a further exploration of the root system (^{Lynch, 2007}) and vigor stem and leaf (^{Guohua et al., 2008}); and the other to have a higher rate of translocation of N from the stem and leaves after formation antithesis to the corn cob (^{Gallais and Hirel, 2004}). The last two authors point out the positive interaction genotype x N impacts increased production associated with increased activity of glutamine synthetase (GS) in the stage of vegetative development (formation of a larger grains number per cob) and that there is negative correlation production grain with early senescence of the photosynthetic apparatus (grain filling).

The profitability of crops is possible with the adoption of best practices fertilization (^{Bruulsema, et al., 2008}; ^{Carneiro et al., 2013}), seeded with higher densities and different topological arrangements (^{De la Cruz et al., 2009}) as narrow grooves (^{Reta et al., 2003}), a double row (^{Gozubenli et al., 2004}) and the combination of two or more above factors with hybrids with "high yield potential" to optimize the use of light, water and nutrients (^{Hodges and Evans, 1990}; ^{Farnham, 2001}; ^{Shapiro and Wortmann, 2006}; ^{Cueto et al., 2006}; ^{Rivera et al., 2007}; ^{Barbieri et al., 2008}; ^{Cervantes et al., 2013}).

The aim of the study was to establish the relationship of grain yield of hybrid corn in Highland valleys State of Mexico with a higher density, annual sequences N fertilization on the ground and define the harvest index (CI) of the macro nutrients N, P, K, Mg and S in grain measured in the total plant. This objective is a continuation of research on N uptake dynamics of vegetative growth to reproductive corn (^{Ciampitti and Vyn, 2011}); performance measurement and extraction of NPK of forage maize in narrow rows (^{Reta et al., 2007}); the fate of nitrogen fertilizer in maize (^{Rimski et al., 2008}); and changing the composition of mineral elements of the corn kernel by "high and low" doses of N (^{Yu-kui et al., 2009}).

Materials and methods

The experiment was planted the years 2009, 2010 and 2011 with the hybrid corn grain white H-51 AE in the first week of May at the Rancho San Nicolás municipality of Timilpan in the State of Mexico with an altitude of 2 628 meters above sea level coordinates 19º 34’ 23.28” north latitude and 99º 46’ 25.62” west longitude. The local climate is classified as temperate sub-humid with abundant rainfall in the months of august to september. In pre sowing the land was watered every year in april with 15 cm sheet of water. The accumulated rainfall from May to November of 808 mm were recorded in 2009 with surplus in september and october; 459 mm well distributed in 2010, and 387 mm in 2011 with drought in july and august more frost in grain filling on 8 and 9 September and then sudden rainstorm interruption. Adding the 150 mm water irrigation tip of each year the rains, maize was calculated received 958, 609 and 537 mm of water in the years 2009, 2010 and 2011, respectively.

Some soil characteristics vertisol pelic analyzed in laboratory according to the ^{SMCS A. C., 1987} and ranked by criteria ^{Alarcón (2004)} were: texture "sandy clay loam" pH of 5.23 units (strongly acidic), CE extract 0.58 dS m^-1 (not saline), and organic matter 1.98% (moderately low). Macro nutrients in mg per kg of dry soil (SS) were measured: inorganic N to 20.9 (mod high.), phosphorus with Bray 1 method of 37. 3 (mod high.), potassium with 198 (mod low.), calcium 1 449 (mod. low), magnesium 454 (mod. high), and 15.8 ppm of sulfur (mod. high). Micro elements in mg kg^-1 of SS: boron with 0.47 (bass), zinc 0.96 (mod low.), copper 80.47 (mod low.), iron 159 ppm (very high), manganese 58.6 ppm (very high) and aluminum 25.2 ppm (very high). It stands the ratio 7.31 Mg/K is calculated on me/100 g SS (high) and Ca/Mg (low).

The design of six treatments (Table 1) evaluated the planting of hybrid H-51 AE with two densities (D1= 65K and D2= 85 000 plants ha^-1) and three annual sequences or the soil (N always N₁= N-N-N, the first year without nitrogen N₂= 0-N-N without three years N₃= 0-0-0) and two doses of N (N= 300 and N= 180 kg ha^-1). The combination of factors 2Di*3Ni= 6 treatments where: T₁= D₁*N₁, T₂= D₁*N₂, T₃=D₁*N₃, T₄=D₂*N₁, T₅= D₂*N₂ and T₆= D₂*N₃. Nitrogen fertilization was supplemented to the ground at planting with P₂O₅, K₂O, MgO, S y Zn. At T₁ to T₃ treatments in the three years they applied the dose in kg ha^-1 of 0-9090-44-50-3. At T₄ to T₆ treatments in the three years 0-4545-0-0-0 formula was applied. The N was fractionated 20% at planting, 40% in V_4-5 with first hoeing and 40% in V_8-9 in second hoeing with "tiro animal" given the height of the plants.

A design with two factors (two densities x three sequences including two doses of N) in completely random arrangement and four replications. The variables measured were: grain yield adjusted to 14% moisture index grain/ cob, index grain/biomass plant, number of cobs ha^-1, grains/cob, grains/ m², weight of 100 grains, and the weight of the dry biomass constant the oven at 65 °C of grain and straw (including stem, leaves and cobs); which they were analyzed separately in the laboratory to measure their contents of N, P, K, Mg and S according to published by ^{Alcántar and Sandoval (1999)} methodologies. Variables and forage grain yield useful plot were measured in six rows of 0.8 m x 10 m long. The remaining variables were measured nine plants random useful plot, values that were added to the performance variables. Hypothesis tests were made with the ^{SAS (1998)} package and comparisons of means by simple effect and interactions were separated with Tukey test at 5% error. Pearson correlation coefficients were calculated of the variables grain production and total biomass against the contents of macronutrients.

Results and discussion

Based on the statistical analysis of three years, there was highly significant response of agronomic variables (Table 2) and the contents of macro elements in the biomass of corn and harvest index on grain, magnesium except on the cob or husk (Table 3). The average grain yield in t ha^-1 was 5.15; and it stands out from the second year of 7.24a with climate "optimal" the worst performance vs 2.68c third year with drought and frost in grain filling. The average rate of grain/biomass was 0.35 in contrast to year 3 with frost and drought sinister 0.25; this is the grain represented a third to a quarter of the total biomass. The variation between cycles production was strongly influenced by variations in climate and management according to ^{Witt et al. (2006)}; ^{De la Cruz et al. (2009)} and ^{Kibet et al., (2009)}.

The total contents of five macro-elements, expressed in kg ha^-1 in the total biomass of corn (grain, straw and cob) on average three years were: 187.9 of N, 30.6 of P 69.9 of K, 24.1 of Mg and 11.6 of S (Table 3). Harvest rates of these elements in corn grain were: ICN= 0.45, ICP= 0.75, ICK= 0.21, ICMg= 0.45 and ICS= 0.43. These values are plausible to those published by the International Institute of Plant Nutrition (IPNI, 2009) for a harvest of 10 t ha^-1 of grain and corn straw respectively; and record: N of 0.66, P of 0.75, K of 0.21 and Mg of 0.28; which are equivalent in that grain harvest amounts in kg ha^-1 of 145N, 30P, 40K and 8Mg. Similarly, these results are consistent with those published by ^{Ciampitti and Vyn (2011)} in relation to nitrogen Ciampitti and García (2007) concerning nutritional requirements associated with absorption and removal of nutrients and studies García (2009) on the efficient nutrient management in maize.

There was significant difference in the average grain t ha^-1 cycles and were: year 1= 5.59b, year 2= 7.24a and year 3= 2.68c (gray stripe data in Table 4). There was also difference in mean grain t ha^-1 treatment effect of density and annual sequence of N and excelled T₁= 7.07a, followed by T2 and T4 with 5.86b and 5.61b, respectively. By increasing the density of plants per ha^-1 to harvest 65 thousand (T₆) to 85 thousand (T₃) of the sequence "always without nitrogen" (00-0); It was obtained from 3.40d to 4.28c t ha^-1 (last column of Table 4). When comparing treatments within each year; grain yields displayed on t ha^-1 sequence 0-0-0 (without nitrogen three years); year 1 with drought/flood occurred T₃= 5.07de vs T₆= 3.69defgh (statistically equal because of high variation data shown numerically but should increase density but not N is applied in this condition); year 2 with optimal climate T₃= 4.61defg vs T₆= 4.68defg (retail cost is decided by lower density if not fertilized with N); year 3 climate with drought and frost T₃= 3.14ghij vs T₆=1.83j again overlooked statistically equal but numerically C.V. appreciated density should be increased but not N is applied in this condition). In summary; if you do not have N fertilization, ensuring greater population yields slightly higher grain according to studies (^{Shapiro and Wortmann, 2006}; ^{Rivera et al., 2007}; ^{Barbieri et al., 2008}; ^{Cervantes et al., 2013}).

Other orthogonal comparisons of this kind are not valid for the other sequences of N and only proceeds to make pairwise comparisons of Di*Ni "high" versus "low" density in the study of three years and within years. Thus, the average grain yield in t ha^-1 in three years of high Di*Ni (T1= 85 thousand plants*300 N always) combination was 7.07a vs 5.61b of Di*Ni low (T4= 65 thousand plants*180 N always). And cycles: year 1. - T₁= 7.61bc and T₄= 7.04c; year 2.- T₁=10.28a and T₄= 7.76bc; and year 3.- T₁= 3.32fghij and T₄= 2.03j. The decision then is: a higher density of plants with annual sequence always high doses of N increases performance; if and only if the weather conditions (second year) occurs with rains in sufficient quantity (or drought or excess first year, or drought and frost filled cob third year) and distributed during the cycle and frost. The availability of moisture and adverse weather factors factor leading to optimize the production system according to ^{Farnham (2001)}; and suggest strategies for genetic improvement in both the efficiency of nutrient absorption (^{Lynch, 2007}) and assimilation (^{Moreno-Pérez et al., 2004}; ^{Guohua et al., 2008}).

What happens when you return to the practice of N fertilization skip to the ground after one year? When comparing average grain in three years in Table 4 of the annual sequence of 0-NN with 300 kg ha^-1 (T₃= 5.86b) vs T₅= 0-N-N with 180 kg ha^-1 (4.80c) shows it increased a ton of grain. By analogy in year 1 with T₂= 5.23d vs T₅= 4.89def (initial drought and excessive rainfall in formation cob); year 2 with T₂= 8.81ab vs T₅= 7.33bc; and year 3 with T₂= 3.55dfghi vs T₅= 2.19hij (drought and frost). Invoked early game theory about risk and investment. First you must ensure sufficient density planting with moderate doses of fertilizer; and as the weather is favorable will be invested in more inputs. Extreme performance data t ha^-1 grain 10.28a of T₁ of the year 2 (D₁, 300 N-N-N) vs 1.8 j T₆ of year 3 support the above statement according to (^{Palafox et al., 2005}; García, 2009).

The means of production of total biomass in t ha^-1 (grain, forage and cob) compared to three-year sequence (gray stripe, Table 5), for treatments density*fertilization (last column) and interactions of the year 1, 2 and 3 (6t set*3 years). Thus, the first year with initial drought and flooding in ear formation was 12.85b, compared to the second year in optimum condition 14.52a and the third with extreme drought and early frost of 9.34c. As for the means of three years treatments Di*Ni high density stands T₁ sequence 300N-N-N with the end 16. 61a and T₆ low density without N three years 8.69. Interactions of years 1, 2 and 3; is valid comparison in columns within one year and to assess treatment sequence of years in horizontal lines and set of 18 combinations of Ti*Ai in either direction. The higher value averages for the year 2 in contrast to depressed values of year 3. More than three times meant biomass in t ha^-1 of T₁/ year 2= 21.14a vs 6.31h of T₆/year 3. The above data match ^{Rivera et al. (2007)} on increased production of forage maize hybrids and more density.

There was highly significant difference of N content in grain and total biomass per cycle (Tables 6 and 7, gray stripes) and these values ICN of 0.45a, 0.51a and 0.35b of years 1, 2 and 3 respectively calculated (Table 8, gray stripe). Similarly the effect of treatment (last columns of Tables 6, 7 and 8); gradual reduction of the content of N in T1 to T6 kg of grain is identified with 122.3a to 52.0d; in biomass 285.2a to 121.7e and ICN there was no difference and the averagewas 0.44. To abound in comparisons of treatments each year, mean differences of the contents of N in grain and total biomass of the first two cycles (year 1 and 2) are observed but were NS in year 3 with drought and frost to fill grain. The N plasticity as to partition in grain and forage tissues by variations in climate and treatments illustrated with ICN consistency of Table 8. These results are consistent with those published by ^{Rimsky et al. (2008)} regarding the fate of nitrogen fertilizer plant corn and grain composition (^{Yukui et al., 2009}).

There was highly significant difference of P content in grain and total biomass per cycle (Table 9) and ICP of these values 0.73b, 0.82a and 0.67c the years 1, 2 and 3 was calculated respectively. Similarly the effect of treatment (last columns of Table 9); gradual reduction of P content in T₁ to T₆ in kg of beans with 34.9a to 13.7d is identified; in biomass 44.7a to 19.4d and ICP difference existed only for the sequence of N 0-0-0 and the average was 0.70 compared to those of IPNI of 2009 of 0.75. These data allow to deduce P phosphorus is linked to the function of N and positively proportional manner; that is, more content of N, more than P. Is consistently above to compare the effect of treatments in years 1 and 2 (year 1 with initial drought and floods to form cob and in the year two optimal) concept, but were NS in year 3 with drought and frost grain filling. Like N, the ICP presented no mean difference effect of treatments within each production cycle.

Finally, positive correlations were found based on Pearson Prob > |r| with H0: Rho= 0, the following pairs of variables: grain at 14% moisture with total biomass (0.9), grain N (0.79), grain P (0.9), grain Mg (0.9), biomass N (0.86), biomass P (0.95), biomass K (0.94) and biomass S (0.87). This is interpreted according to IPNI, 2009; plant nutrition is associated with increased forage production and this with the corn grain yield.

Conclusions

The absolute value of the contents of macro nutrients in the total biomass and grain maize hybrid H-51 AE were higher associated with optimal weather condition (year 2), higher density (85 000 plants ha^-1) and increasing doses of N to the ground, as well as additional fertilization for elements magnesium and sulfur. The total contents of five macroelements, expressed in kg ha^-1 in the total biomass of corn on average three years were: 187.9 of N, 30.6 of P, 69.9 of K, 24.1 of Mg and 11.6 of S. these values to corn or grain harvest indexes of these elements were: ICN= 0.45, ICP= 0.75, ICK= 0.21, ICMg= 0.45 and ICS= 0.43; and these were affected only by the age factor associated with climate. When referring to the year 2 optimal climate as "normal", the value of ICN third year found (initial and frost drought in formation cob) was reduced and a slight reduction in ICP of year 1 (drought/flood) and drastic reduction in year 3 (drought/frost). Increasing planting density 85 000 plants ha^-1 meant slight increases in grain yields in years of adverse weather and very significant in optimal climate. The N fertilization of the ground should be held in every year and may be more decisions associated with prognosis and occurrence of climate in Highland valleys.