Bromatological evaluation of eleven corn cultivars harvested at two cutting heights

The objective of this study was to evaluate the chemical composition and dry matter in vitro digestibility of stem, leaf, straw, cob and kernel fractions of eleven corn (Zea mays) cultivars, harvested at two cutting heights. The experiment was designed as randomized blocks, with three replicates, in a 2 × 11 factorial arrangement (eleven cultivars and two cutting heights). The corn cultivars evaluated were D 766, D 657, D 1000, P 3021, P 3041, C 805, C 333, AG 5011, FOR 01, CO 9621 and BR 205, harvested at a low cutting height (5 cm above ground) and a high cutting height (5 cm below the first ear insertion). Cutting height influenced the dry matter content of the stem fraction, which was lower (23.95%) in plants harvested at the low, than in plants harvested at the high cutting height (26.28%). The kernel fraction had the highest dry matter in vitro digestibility (85.13%), while cultivars did not differ between each other. Cob and straw were the fractions with the highest level of neutral detergent fiber (80.74 and 79.77%, respectively) and the lowest level of crude protein (3.84% and 3.69%, respectively). The leaf fraction had the highest crude protein content, both for plants of low and high cuttings (15.55% and 16.20%, respectively). The increase in the plant cutting height enhanced the dry matter content and dry matter in vitro digestibility of stem fraction, but did not affect the DM content of the leaf fraction.


Introduction
The achievement of higher productivity levels in cattle breeding systems requires the use of high quality food.In tropical regions, the seasonality of forage production is a major problem for obtaining high productivity, so the roughage supplementation becomes necessary during the dry season.
Traditionally, due to its high nutritive value for ruminants, mostly in terms of energy, corn (Zea mays) is the most commonly used plant for silage production.In addition, corn has all the necessary characteristics to a proper fermentative process inside the silo, such as adequate levels of dry matter (DM), soluble carbohydrates and buffering capacity.Generally, cultivar, maturity and mechanical processing also influence the chemical and physical characteristics of the carbohydrates in the silage (Johnson et al., 2002;Velho et al., 2010).
When harvesting corn cultivars for silage production, cutting height is one of the factors that can affect both production and nutritive value of the silage (Neumann et al., 2007).The variability in production and quality parameters observed among cultivars has been encouraging research aiming at the knowledge of the performance and identification of the best varieties and/or hybrids (Molodo et al., 2010;Coimbra et al., 2010).When compiling the results of eleven studies in which the corn for silage production was harvested at different cutting heights, Wu & Roth (2005) observed that when the corn was cut high (leaving about 50 cm of stalk in the field), the levels of crude protein (CP) and energy, as well as neutral detergent fiber (NDF) digestibility and milk production by silage ton enhanced, compared with conventional cutting height (about 17 cm above the ground).Wu et al. (2001) and Neylon & Kung (2003) also reported an increase in milk production, when cows were fed silages made from corn cut at higher cutting heights; however, both studies reported a decrease in milk fat percentage.
Knowledge of kernel percentage and composition in the ensiled mass, as well as of the different components of the plant as a whole, is very important for corn silage production and for measuring its implications in the final nutritional value of the food.In this regard, Nussio & Manzano (1999) suggest that in cultivar selection programs for silage production, predictive models of silage quality be established based on two factors: kernel percentage in the ensiled mass (% of DM) and nutritional value of stalk and leaf fractions (% of true in vitro DM digestibility %).
In this way, the objective of this research was to evaluate changes in the bromatological characteristics of different corn plant fractions according to cutting height.

Material and Methods
This study was conducted in the experimental area of the Fazenda de Ensino, Pesquisa e Produção da Faculdade de Ciências Agrárias e Veterinária (FCAV) -UNESP, Jaboticabal campus.The experiment was designed as randomized blocks (three blocks), in an 11 × 2 factorial arrangement (eleven cultivars and two cutting heights: high and low).However, because of the fact that cutting height did not alter these parameters, the chemical characteristics of straw, cob and kernel fractions were analyzed considering three blocks with two replicates each, and eleven treatments (cultivars).The height of the first ear was analyzed considering three blocks and eleven treatments (cultivars), because this variable was determined only in plants harvested at the low cutting height.
The experimental area was conventionally prepared with deep moldboard plowing (30 cm), followed by two harrow plowings immediately before sowing.Plots were composed of six 5-m rows at 0.9-m row spacing.Ten seeds per meter, two by two, spaced by 20 cm in a row, were hand sown over fertilized furrows, with the aid of a jab planter.After emergence, the number of plants was reduced to five plants per linear meter, by manual thinning, in order to obtain a final population of 55,000 plants per hectare.Cover fertilization was performed 25 days after seed germination.
Plants were harvested at two cutting heights: low (5 cm above ground) and high (5 cm below first ear insertion).Harvesting occurred when, by visual evaluation, the milkline of the central ear kernels was at approximately 2/3 of the kernel, approximately 95 days after planting.Harvesting occurred when, by visual evaluation, the milk-line of the central ear kernels was at approximately 2/3 of the kernel.Plant fractions were manually separated.
To each fraction (stalk, leaf, straw, cob and kernel) levels of crude protein (CP), ether extract (EE), ash, neutral detergent fiber (NDF), acid detergent fiber (ADF) and cellulose were determined according to Silva & Queiroz (2002).Cellulose was determined by lignin oxidation with potassium permanganate, whereas lignin was considered as the difference between ADF and cellulose concentrations.
Hemicellulose, in turn, was determined as the difference between NDF and ADF concentrations.
No antifoam agent was used for the NDF and ADF determination, and no sodium sulfite was used for the NDF determination.However, samples rich in starch (kernels) were added 0.2 mL of amylase PA (Ankom).
Samples were weighed in proper digestion bags and put into the digestion recipient containing the solutions and the ruminal fluid inoculum.The ruminal fluid inoculum was obtained from a rumen cannulated Holstein steer that was adapted for 15 days in feedlot, receiving about 25 kg corn silage, 2 kg concentrate and water ad libitum.During the ruminal fluid collection period, the steer received the same diet.
Results were submitted to analysis of variance and means were compared by Tukey test at 5% probability.Statistical analyses were performed using SAS statistical package (1999), and all variables were tested for residue normality.

Results and Discussion
There was no difference in the ether extract and crude protein levels of stalk fraction (P>0.05) for cultivars according to cutting height (Table 1).However, cultivars harvested at the high cutting height had higher dry matter and ash levels (P<0.05),compared with the low cutting height.
Dry matter of talk fractions levels of cultivars C 805 (25.97%) and AG 5011 (21.63%), harvested at the low cutting height were greater than those observed by Almeida et al. (2003), 21.86% and 16.22%, respectively.On the other hand, stalks from cultivars AG 5011 (21.63%) and P 3041 (24.34%) showed dry matter levels similar to those observed by Zopollatto et al. (2009), Beleze et al. (2003) and Bernard et al. (2004), who cut the plants at 116 and 118 days after sowing, respectively.The concentration of ash did not differ between cultivars Dina 766, Dina 1000, P 3041 and AG 5011, which had higher ash concentration in the stalk fraction (P<0.05)than the other evaluated cultivars (Table 1).
Cultivars Dina 766 and AG 5011 had higher crude protein levels in the stalk fraction than cultivar C 805, similarly to the results observed by Almeida et al. (2003): 4.21 and 2.36% respectively.Crude protein levels in the stalk fraction did not differ between cultivars Dina 657, Dina 1000, P 3021, P 3041, C 805, C 333, AG 5011, FO 01, Dina Co 96 21 and BR 205.
There was interaction effect between cultivar and cutting height for the dry matter in vitro digestibility (DMIVD) coefficients of the stalk fractions (P<0.05)(Table 1).The DMIVD differed between cultivars (P<0.05).Cultivar Dina 657, when cut low, had lower DMIVD than when cut high (47.68 and 61.55, respectively).According to Paziani et al. (2009), plant digestibility depends mostly on the stalk digestibility and on kernel related parameters, both influenced by cutting height.
The neutral detergent fiber concentration of stalk fraction was not influenced by cutting height (P>0.05).However, plants cut high had higher hemicellulose levels than the ones cut low (P<0.05).This result may be explained by the higher acid detergent fiber levels observed in the stalks evaluated (Table 2).There was interaction between cultivar and cutting height for the concentrations of ADF, cellulose and lignin in the stalk fraction (P<0.05)(Table 2).Kung Junior et al. (2008) observed a 9% reduction in the ADF concentration as cutting height increased.According to the authors, this may be explained by the lower contribution of the stalk fraction observed in silages with lower ADF content.Kruczynska et al. (2001) observed a reduction in ADF levels and greater effective digestibility of the silage when plants were harvested at 50 cm of height, compared with when plants were harvested at 10 cm of height.Considering the low cutting height, the stalk fraction of cultivars Dina 766 and FO 01had the highest ADF levels.On the other hand, considering the high cutting height, the extreme values were observed in cultivars P 3041 and C 805; and BR 205, AG5011 and Dina 766, which had the highest and lowest ADF levels, respectively.
Except for cultivar P 3041, cutting height did not affect lignin content of the stalk fraction (P>0.05).Also, cutting height did not alter the ADF or cellulose content of the stalk fraction of cultivars Dina 1000, P 3021, P 3041, C 805 and Dina co 9621 (P>0.05).However, ADF and cellulose concentration of the stalk fraction of cultivars Dina 766, AG 5011, FO 01 and BR 205, harvested at the low cutting height, were significantly higher (P<0.05)(Table 2).
Dry matter, NDF and EE content (Table 3), as well as the DMIVD coefficient (Table 4), of the leaf fraction did not differ between cultivars (P>0.05).On the other hand, ash, ADF, hemicellulose, cellulose and CP levels were significantly different according to cultivar (P<0.05)(Table 3).
Cutting height did not alter (P>0.05) the DM, ash, NDF, hemicellulose or cellulose concentrations of the leaf fraction (Table 3).However, EE and CP levels (Table 3), as well as DMIVD coefficients (Table 4), of the leaf fraction were higher (P<0.05) for plants harvested at the high cutting height, compared with the low cutting height (Caetano et al., 2010).
There was no difference in the NDF concentration of the leaf fraction between cultivars (Table 3).Cultivar Dina 1000 had higher ADF and cellulose concentration in the leaf fraction than cultivar Dina 766 (Table 3).
Cutting height did not alter DM, ash, NDF, hemicelluloses or cellulose concentrations of the leaf fraction (P>0.05).However, the ADF content in the leaves was increased in plants cut high, compared with plants cut low (Table 3).Lignin concentration in the leaf fraction was also not affected by cutting height (Table 4).
Cultivar Dina 657 showed the higher DM content (32.59%) in the straw fraction (25.73%) than cultivar Dina 1000 (Table 5) (P<0.05).However, as both had similar DM levels in the kernel fraction (  There was no difference in the ADF and cellulose concentration (Table 5), as well as for the IVDMD coefficients, of the straw fraction between cultivars (P>0.05).Lignin levels of cultivars Dina 657 and C 805 showed the highest and the lowest values, respectively, differing from each other (P>0.05).
Kernel fraction IVDMD coefficients did not differ between cultivars (P>0.05)(Table 4).The kernel fraction DM concentration ranged from 41.42 to 49.92% (cultivars AG 5011 and C 805,respectively).Generally, these results indicate that harvest based on the milk line evaluation occurred at the proper time.According to Zopollatto et al. (2009), Kennington et al. (2005), and Mello et al. (2005), at ensiling, stalk and kernel are the components with biggest participation, and kernels are the fraction which mostly contributes to the corn silage final quality, since it has higher IVDMD.Cultivars AG 5011 and Dina 766 showed higher ash concentration in the kernel fraction than cultivar P 3041 (Table 7).Cultivars Dina 766 and FO 01 had higher EE concentration in the kernel fraction than cultivars C 333 and BR 205, and cultivar FO 01 had higher CP levels than cultivars P 3021, P 3041 and C 805 (P<0.05)(Table 7).Rosa et al. (2004), evaluating the agronomic profile of corn hybrids for ensiling, when kernels were between the end of the pasty and the beginning of the farinaceous stages (35.5% DM), observed DM levels of 41.0 to 59.3% in the cobs and kernels, respectively, in hybrid AG 5011.
There was no difference in lignin levels and IVDMD coefficients in the kernel fraction (P>0.05) between cultivars (Table 4).However, the NDF and hemicellulose concentration were higher for cultivar C 333, compared with the others (P<0.05).Concentration of ADF and cellulose did not differ in the kernel fraction of the evaluated cultivars (P>0.05)(Table 7).Generally, the kernel fraction had a high IVDMD coefficient and low levels of all the cell wall components.

Conclusions
Increasing corn plants cutting height elevates dry matter level and in vitro dry matter digestibility of the stalk fraction, but does not interfere with the dry matter concentration of the leaf fraction.Among all the studied plant fractions, kernel and cob has the highest dry matter concentrations, and kernel has the highest in vitro dry matter digestibility.

Table 1 -
Bromatological composition and in vitro dry matter digestibility (IVDMD) of the stalk fraction Means followed by the same uppercase letters (row) and lowercase letters (column) do not differ (P>0.05) by Tukey test; P = significance level; CV = coefficient of variation.Means followed by the same uppercase letters (row) and lowercase letters (column) do not differ (P>0.05) by Tukey test; P = significance level; CV = coefficient of variation.

Table 2 -
Bromatological composition of the stalk fraction

Table 7
Means followed by the same uppercase letters (row) and lowercase letters (column) do not differ (P>0.05) by Tukey test; P = significance level; CV = coefficient of variation.

Table 3 -
Bromatological composition of the leaf fraction difference probably occurred due to different water loss rates of the straw fraction.Cultivars P 3041 and Dina 766 had higher ash concentration in the straw fraction, compared with cultivar P 3021 (Table5) (P<0.05).Cultivar P 3041 had higher EE concentration in the straw fraction than cultivar C 333 (P<0.05).Crude protein levels of the straw fraction from cultivar FO 01 were higher, compared with cultivars Dina 657, Dina 1000, P 3021, C 805 and BR 205 (P<0.05).

Table 4 -
Lignin and in vitro dry matter digestibility (IVDMD) of fractions leaf, straw, cob and kernel Means followed by the uppercase letters (row) and lowercase letters (column) do not differ (P>0.05) by Tukey test; P = significance level; CV = coefficient of variation.

Table 5 -
Rosa et al. (2004)position of the straw fraction levels observed in the kernel fraction (Table7), indicating that harvest was done at similar maturation state, and suggests a likely difference in the water loss speed of the cob fraction.Rosa et al. (2004)evaluated the agronomic profile of corn hybrids at the ensiling point (29.4 to 35.5% DM) and verified DM levels of 32.1 to 41.0% and 42.1 to 59.3% for cob and kernel fractions, respectively.The highest levels of ash and CP of the cob fraction were observed in cultivar FO 01 (P<0.05)(Table6),compared with cultivars P 3021 and C 333. Cultivars Dina 766, Dina 657 and BR 205, had higher IVDMD coefficients in the cob fraction than cultivar C 333 (P<0.05).The cob fraction IVDMD coefficients of cultivars AG 5011 and C 805 did not differ between each other (P>0.05)(Table

Table 7 -
Bromatological composition of the kernel fractionMeans followed by the same uppercase letters (row) and lowercase letters (column) do not differ (P>0.05) by Tukey test; P = significance level; CV = coefficient of variation.

Table 6 -
Bromatological composition of the cob fraction