Corn yield , forage production and quality affected by methods of intercropping corn and Panicum maximum

The objective of this work was to evaluate methods of intercropping corn and Panicum spp. forages and their effects on corn yield and forage development. Two experiments with Panicum spp. were conducted, one with 'Tanzania' and other with 'Massai' in the municipality of Piracicaba, in the state of São Paulo, Brazil, in the 2012/2013 growing season. In both experiments, the treatments were: broadcast seeding of Panicum spp. at the same time of corn sowing; corn with Panicum spp. between rows; Panicum spp. sown with fertilizer; Panicum spp. sown between corn rows after corn establishment; broadcast seeding of Panicum spp. after corn establishment; and sole corn sowing. For corn, leaf nitrogen content, ear height, and grain yield were evaluated. For Panicum spp. cultivars, height, dry mass, tiller density, and leaf:stem ratio were evaluated. The intercropping establishment methods used do not affect corn growth, grain yield, and N leaf content. The seeding of both cultivars of Panicum spp. when corn plants had four expanded leaves reduces forage dry mass production and increases the leaf:stem ratio. The Panicum spp. broadcast method, performed after corn was established, does not allow appropriate establishment, with few plants in the area.

Different ways of deploying ICL can be used (Borgui et al., 2013).However, most farms in South America have cultivated the fields with annual crops for years, then, moved on to intercropping between maize and tropical grass (Borgui et al., 2013).After the harvest of the annual crop, fields can be grazed by the livestock (Crusciol et al., 2015).
Panicum spp.have a larger forage production compared to Urochloa species (Silva et al., 2016).This is an important feature to animal grazing after corn harvest in ICL systems, since a larger number of animals can graze the field.In the other hand, Panicum spp.sward erected leaves structure, and bigger height (Mello & Pedreira, 2004) may pose a problem to interspecific competition.If Panicum spp.species were not suppressed by corn, surely corn grain yield would be lower at intercropping.
Considering this, an important issue is the intercropping establishment method between corn and Panicum spp.cultivars.Due to differences of seeders and farm sizes, tropical grasses sowing have been done before, simultaneously or after corn sowing (Borgui et al., 2013).The intercropping between corn and palisadegrass showed that grass sowing simultaneously or after corn has no effect on grain yield (Borghi et al., 2013), with the right herbicide management for this system (Ceccon et al., 2010).However, the pasture production was decreased by 20% when tropical grass was sown after corn planting (Borghi et al., 2013).So, more information about corn and Panicum spp.grass intercropping is necessary.Doubts about grain yield, N uptake, as well as about the perennial crop forage biomass yields and biomass quality should be elucidated.
The objective of this work was to evaluate the most common methods of intercropping systems with Panicum spp.cultivars and their effects on grain yield and grass development.

Materials and Methods
Two independent experiments were conducted in 2012/2013, in municipality of Piracicaba, state of São Paulo, Brazil, 22°41'31"S, 47°38'00"W and 580 m of altitude.Precipitation and temperature of the experimental period are presented on Figure 1.Intercropping between corn and Panicum spp. was established on a Typic Hapludults soil (USDA, 2006).
In both experiments, intercropping methods were compared between 6 treatments: Panicum spp.broadcast seeding aboveground 1 hour before corn sowing (S. broadcast); simultaneously plant sowing with Panicum spp.seeds distributed in furrow between corn rows (S. between rows); simultaneously plant sowing in same row with Panicum spp.seeds in corn fertilizer depth (S.Fertilizer); Panicum spp.seeds distributed in furrow between corn rows concomitantly with top dressing fertilization at 4-leaf corn stage (T. between rows); Panicum spp.broadcast sowing with top dressing fertilization at 4-leaf corn stage (T.broadcast); sole corn cultivation.Experimental design for both trials was randomized blocks with 4 replications.Plots had 3.6 m (4 corn rows) per 10 m length and the total area of each trial was 864 m².
Both Panicum spp.cultivars were sown at the density of 2.4 kg ha -1 of viable seeds in all treatments.To S. between rows and T. between rows treatments, the seeds were sown in a furrow with 0.03 m depth to simulate the seeder or the conventional fertilizer spreader.In S. fertilizer treatment, adjacent furrows to corn rows were made with 0.06 m depth.This procedure simulated what occurs when forage seeds are mixed with corn fertilizer and sown in the same operation; on the other hand, eventual salinity provided by seedfertilizer contact was not included in this research.
The hybrid AG 8061 PRO was sown in both experiments on 12th of December, 2012, with 0.9 m row spacing, aiming at 60,000 plants ha -1 .The sowing fertilizer consisted of 27; 87; and 50 kg ha -1 of N, P 2 O 5 and K 2 O, respectively.At corn stage of four expanded leaves, topdressing fertilization was applied (100 kg ha -1 of N and 30 kg ha -1 of K 2 O).
Corn leaf sampling for N concentration analysis was done when 50% of the corn plants were in full flowering stage.Corn leaf sampling was randomized by choosing 5 plants per plot as methodology proposed by Malavolta et al. (1997).N concentration in leaf samples was determined using methods described by Malavolta et al. (1997).
Just prior to crop harvest, the corn first ear insertion height was measured in two central rows of each plot.Grain yield was determined harvesting ears of 4 m long lines from two central rows.Threshed grains were weighted, the moisture content was determined, and the data was transformed to grain yield per hectare (130 g kg -1 wet basis).
In order to evaluate Panicum spp.cultivars 'Tanzania' and 'Massai', tiller density was measured by counting a number of tillers inside a 0.25 m² area in the center of each plot defined by a frame.Dry mass production was determined by cutting the grass inside a frame of 1 m².The leaf:stem ratio was measured with a subsample, which was dissected into leaves, stem and dead material.All the material was dried in a ventilated laboratory oven at 65ºC for 72 hours for dry mass determination.All samples were obtained on the same day of the corn harvesting.
Data was submitted to test for normality and homogeneity of variance.If these requirements were confirmed, an analysis of variance would be applied to verify if the effects were caused by the intercropping methods.Means were compared by Fisher test (LSD) p≤0.05.The statistical analysis was done using SAS software (SAS Institute Inc., Cary, NC, USA).

Results and Discussion
The methods for establishing grasses did not affect the height of ear -which was in average 1.25 m for both experiments -, grain yield (Table 1) or N concentration in leaves (Table 2).
The height of the corn ear insertion is important as a morphologic change induced by low intensity or different quality of light (Rajcan & Swanton, 2001).Therefore, in this study, it was possible to infer that grasses do not impair corn growth.Borghi et al. ( 2013) observed no differences in height of ear when two methods for intercropping corn and tropical grasses were compared to corn monoculture.Grain yield was the same for intercropped treatments and sole corn cultivation in both experiments and it was, in average, 7,218 kg ha -1 for 'Tanzania' cultivar and 6,540 kg ha -1 for 'Massai' (Table 1).These observations prove the viability of intercropping corn and Panicum spp.grasses in crop and livestock integration systems as reported to other kind of tropical grasses (Baldé et al., 2011;Borghi et al., 2013;Ceccon et al., 2013).
Corn leaf N content at flowering was not affected by the method for establishing grasses, which was in average 34 g kg -1 for both experiments (Table 2).N concentration in leaves was higher than the level considered adequate, which is 27.5 to 32.5 g kg -1 (Malavolta et al., 1997).Therefore, it is presumed that the cultivation of Panicum spp.grasses with corn did not reduce the amount of available N in soil to affect N uptake by corn plants.Crusciol et al. (2013) also found no difference in leaf N content when sole corn cultivation was compared to intercropped corn and Urochloa sp.
Height, dry mass production, tiller density and leaf:stem ratio of the Panicum spp.cultivars were affected by intercropping methods (Tables 3 and 4).Both cultivars ('Tanzania' and 'Massai') had the same response to the different methods.Grasses sown right before or during the corn seeding (S. broadcast, S. between rows and S. fertilizer) reached greater height and produced more dry mass as well (Table 3).
Panicum spp.height, when sown after corn (T. between rows and T. broadcast), was lower (51% for 'Tanzania' and 56% for 'Massai') than in establishing methods on the same day of corn.The dry mass production was 3,995 kg ha -1 for 'Tanzania' and 1,638 kg ha -1 for 'Massai' when established on the same day of corn; and 479 kg ha -1 for 'Tanzania' and 282 kg ha -1 for 'Massai' when the grass was sown in furrow after corn germination (T. between rows), 86% lower than in the first treatments.On the other hand, the treatment T. broadcast produced only 85 kg ha -1 and 27 kg ha -1 of dry mass, for 'Tanzania' and 'Massai' respectively, and it was the less productive treatment (Table 3).
Panicum spp.dry mass production, when it was sown together with corn, was higher than in topdressing treatments because the initial growth of the grasses was prior to the shading by corn plants, and thus, the grasses had more incidence of light and development time on the beginning.Ferreira et al. (2010) reported shading affecting Panicum spp.dry mass production, and Borghi et al. ( 2013) observed higher dry mass production of tropical grasses when seeded simultaneously with corn compared to later sowing.
Low dry mass production, when Panicum spp.cultivars were broadcast sown at topdressing fertilization (T.broadcast), is explained by the lower plant population.This stablishing method provided 60  and 147 tillers m -2 for 'Tanzania' and 'Massai' grasses respectively, 71% or 64% less than the average of other treatments, which had 211 tillers m -2 for 'Tanzania' and 412 tillers m -2 for 'Massai' (Table 4).Grass seed incorporation in soil is crucial for good plant establishment (Freitas et al., 2005).Therefore, it is not recommended to establish the intercropping with broadcast seeding after the corn plant development.
For the other treatment of broadcast seeding of Panicum spp.before corn (S. broadcast), this fact did not occur because the seeder, when working after grass seed distribution, provided the necessary seed incorporation.
The leaf:stem ratio data was higher when grasses were sown after corn germination (T.broadcast and T. between rows).When the grasses were sown simultaneously with corn (S. broadcast, S. between rows and S. fertilizer), the ratio was 64% lower than in the other methods for 'Tanzania' and 61% lower for 'Massai' (Table 4).The earlier grass establishment in S. broadcast, S. between rows, and S. fertilizer influenced the pattern of growth.In such cases, the grass developed greater height and consequent lower leaf:stem ratio.Several studies positively correlate height growth of grasses with elongation of stems and consequent lower leaf:stem ratio, which is attributed to competition between tillers per light (Sbrissia & Silva, 2008).Paciullo et al. (2011) observed higher stem elongation rate in three species of Urochloa sp. by decreasing the intensity of available light.
Recommended pre-grazing height of 'Tanzania' pasture is 0.75 m, which corresponds to 95% of sunlight interception and higher dry mass production of leaf lamina associated with the control of stem elongation (Zanine et al., 2011).'Tanzania' height, when seeded simultaneously with corn, was 1.51 m, which demonstrates that the grass exceeded the management goal because of the delay in its harvest due to the presence of the corn.
Treatments in which the grasses were sown after corn obtained suitable values of height, 0.76 m and 0.56 m for 'Tanzania' and 'Massai' respectively, and leaf:stem ratio 1.5 for 'Tanzania' and 2.9 for 'Massai' (Tables 3  and 4).This happened due to shorter development time during intercropping.Zanine et al. (2011) obtained leaf:stem ratio of 2.78 also with 'Tanzania' grass, adopting pre-grazing heights corresponding to 0.75 m.
The elevated grass height related to the low quantity of dry mass of the 'Tanzania' and 'Massai' can also be explained by the greater influence of corn shading throughout development (Portes et al., 2000).The average height of 'Tanzania' grass sown on the same day as corn was 1.51 m whilst 0.8 m when sown after corn germination.The dry mass observed was only 3,995 kg ha -1 for the methods for establishing on the same day as corn and 479 kg ha -1 when 'Tanzania' was sown in furrow after corn.In an experiment in which 'Tanzania' cultivar was grazed at 0.75 m, Zanine et al. (2011) obtained the herbage pre-grazing mass of 6,070 kg ha -1 .
The methods S. broadcast, S. between rows and S. fertilizer produced dry mass for grazing cattle in an integrated crop-livestock system, or for covering the soil for no-tillage cultivation of annual crops for the next growing season.Meanwhile, the long development period of the grass in these treatments favors the production of stems relative to leaves, especially in the case of the cultivar Tanzania.According to Valente et al. (2010), the leaf:stem ratio is positively correlated with digestibility of grass; and the high proportion of stem in the sward can limit animal intake (Gontijo Neto et al., 2006).The 'Massai' cultivar showed a lower proportion of stem, making it more suitable for simultaneous establishment with corn.T. between rows produced smaller amount of dry mass at corn harvest when compared to the simultaneous establishing methods of intercropping, and it can be used for soil cover or grazing in a period after corn harvest.According to Zhang & Li (2003), grasses have additional growth after the main crop harvest.

Conclusions
1.The intercropping establishment methods used do not affect growing, grain yield, and N leaf content of corn when it is intercropped with both Panicum spp.cultivars.
2. The seeding of both cultivars of Panicum spp.when the corn plants had 4 expanded leaves reduces forage dry mass production, and increases leaf:stem ratio.
3. The broadcast seeding of the forage simultaneously to top dressing fertilization method is not feasible on the experimental conditions because the grass is not properly established, with few plants in the area.

Table 1 .
Height of ear (m) and grain yield with 13% of humidity (kg ha -1 ) in an intercropping system between corn and Panicum spp.
(1) S. broadcast: Panicum spp.broadcast seeding before corn sowing; S. between rows: Panicum spp.seeds distributed in furrow between corn rows; S. fertilizer: Panicum spp.seeds in corn sowing fertilizer depth; T. between rows: Panicum spp. in furrow between corn rows at top dressing fertilization; T. broadcast: Panicum spp.broadcast sowing with top dressing fertilization; Sole corn: corn without Panicum spp.ns Nonsignificant by the F test.

Table 2 .
Corn leaf N content intercropped with Panicum spp. at the corn flowering.
fertilizer: Panicum spp.seeds in corn sowing fertilizer depth; T. between rows: Panicum spp. in furrow between corn rows at top dressing fertilization; T. broadcast: Panicum spp.broadcast sowing with top dressing fertilization; Sole corn: corn without Panicum spp.ns Nonsignificant by the F test.
row between corn rows; S. fertilizer: Panicum spp.seeds in corn sowing fertilizer depth; T. between rows: Panicum spp. in furrow between corn rows at top dressing fertilization; T. broadcast: Panicum spp.broadcast sowing with top dressing fertilization.***Significant at less than 0.1% probability of error by the F test.

Table 4 .
Tiller density (number m -²) and leaf:stem ratio of Panicum spp.cv.Tanzania and cv.Massai intercropped with corn(1).Means followed by the same letter in the columns do not differ by LSD test at 5% probability. (2)S. broadcast: Panicum spp.broadcast seeding before corn sowing; S. between rows: Panicum spp.seeds distributed in furrow between corn rows; S. fertilizer: Panicum spp.seeds in corn sowing fertilizer depth; T. between rows: Panicum spp. in furrow between corn rows at top dressing fertilization; T. broadcast: Panicum spp.broadcast sowing with top dressing fertilization; Sole corn: Corn without Panicum spp.**Significant at less than 1% probability of error by the F test.