The study was carried out in an experimental area located at the School of Agronomy of the Federal University of Goiás, in the city of Goiânia, Goiás. The following average climatic indicators were verified for the locality: annual precipitation of 1575 mm and average monthly temperature of 22.9 °C, predominance of Aw climate, characterized by tropical climate with rainy season between October and April and a period with precipitations below 100 mm monthly between May and September (Cardoso et al., 2014). Climatic records of air temperature and relative humidity, occurring during the conduction of the crambe crop, can be observed in Figure 1.

Figure 1

The soil was classified as RED LATOSOL (oxisol), following a methodology proposed by Embrapa (Santos et al., 2013). The chemical analysis of the soil before sowing of the cover crops showed, at depth of 0-0.2 m, levels of de Ca²⁺: 2.0 cmol_c dm^-3, Mg²⁺: 0.81 cmol_c dm^-3, K⁺: 0.34 cmol_c dm^-3, P (Mehlich I): 3.5 mg dm^-3, organic matter: 7.0 g dm^-3, Al³⁺: 0.0 cmol_c dm^-3, H+Al: 2.5 cmol_c dm^-3 and pH (CaCl₂): 4.6, CTC: 5.6 cmol_c dm^-3, V%: 55.7%, according to the procedures recommended from Donagemma et al. (2011). The granulometric analysis of the soil presented 48 g kg^-1 of clay in the depth of 0-0.2 m following the procedures in Silva (2009).The experimental design was a randomized block design, with six treatments containing cover crop residues of the following cover crops: crotalaria (Crotalaria juncea L.), pigeonpea (Cajanus cajan(L.) Hunth), jack bean (Canavalia ensiformis (L.) DC.), millet (Pennisetum glaucum (L.) R. BR.) and sorghum (Sorghum bicolor (L.) Moench), besides a control treatment, with the use of spontaneous vegetation, with a predominance Ipomoea quamoclit L., Bidens pilosa L. and Amaranthus viridis L., with four replications. Each experimental plot had dimensions of 0.8 m x 1.20 m.

The cover crops were sown on December 08 of 2015, except for the spontaneous plants, which were naturally emerged by the seed bank itself in the plots corresponding to the control treatment. At 50 days after sowing the cover crops were desiccated with herbicide application (Glyphosate, 360 g L^-1 of active ingredient) and at 22 days after application of the herbicide, a cut was made for the deposition of the straw on the soil. The straw was distributed over the area corresponding to the plots.

Prior to sowing of the crambe, cultivation of curly lettuce was carried out in a no-tillage system for 50 days. The crambe was sown on June 3 of 2016, in open furrows with a hoe, distant 25 cm between each other and with depth of three centimeters. The seeds were distributed in order to establish a final stand of 23 plants m^-2.

Irrigation was performed daily through hoses with drippers spaced 20 cm apart between the lines. Through this system was also applied the cover fertilization, applying 40 kg ha^-1 of nitrogen and potassium, divided in the four weeks that followed the emergence of the plants. During crambe cultivation there was no necessity of pests control, diseases or spontaneous plants, justifying the non-use of agrochemicals for these purposes.

A first evaluation was performed 49 days after sowing, when the culture entered the reproductive stage. At this moment, the plant height parameters were evaluated, measuring since the lap of the plant until the apical end of the main branch with a ruler, stem diameter at the soil level with digital caliper (Metrotools, São Paulo, SP, Brazil), number of leaves, relative leaf chlorophyll content, by reading with a hand-held chlorophyll meter (CFL1030; Falker, Porto Alegre, RS, Brazil), number of lateral branches, fresh and dry biomass of aerial part. The shoot, root and total dry mass were measured in a digital scale (ML 600, Marte, São Paulo, SP, Brazil) after drying the plants in an oven at 65 °C until obtaining constant mass.

At 95 days after planting, the grains of three plants per plot were harvested. The humidity was adjusted to 13% to determine the grain mass per plant, the hectoliter weight of the grains and the productivity of each treatment. The gross revenue of each treatment for one hectare was also estimated. For this, was used the value of R$ 400.00, paid per ton of grain by the industry (Roscoe et al., 2010). Data were submitted to analysis of variance and compared by Tukey test at 5% probability.

The averages obtained for of height, stem diameter, number of leaves and branches were 62.11 cm, 15.95 mm, 70.5 and 16.9, respectively. These results may be related to the good condition of the soil, even under the control treatment, due to the effect of the cover crops on the addition of organic matter to the soil, nutrient cycling, physical protection and soil conditioning by root distribution. Previous cultivation of the lettuce, in which cover fertilization was carried out, also improved the soil chemical conditions in the area. It was verified that, for the biometric variables of height, stem diameter, number of leaves and branches, there were no significant statistical differences (Table 1).

Table 1

For the relative contents of leaf chlorophyll, it was observed superiority of plants cultivated on the residues of crotalaria, followed by millet. However, treatment differed significantly only from the control (Table 2). This result is related to the better nutritional conditions of the plants, due to the qualities of the straw. Torres et al. (2005) observed that the crotalaria and millet crops accumulate large amounts of nitrogen, which is released after straw formation. According to Cazetta et al. (2005), millet is also viable in crops aiming at dry matter production, reaching more than 10 t ha^-1, contributing to improve soil physical conditions. In broccoli (Brassica oleraceae), Perin et al. (2004) observed that the presence of crotalaria increases the content and accumulation of N in the leaves and inflorescences, both in the absence and presence of N-fertilizer.

Table 2

Studying the effect of the different plant species biomass presence on soil microbial biomass activity, Carneiro et al. (2008) observed that fallow development with spontaneous plant development, common practice in several regions, impairs the sustainability of the soil by causing detriments in the carbon of the microbial mass and in the organic carbon of the soil. On the other hand, the presence of the pigeon pea, C. juncea, niger and oats provided increases between 93% and 190% in the carbon of the microbial biomass. The presence of cover crop residues in the present study probably maintained high levels of organic carbon in the soil, thus favoring the development of crambe.

The microbial biomass of the soil has great importance as a catalyst for the decomposition process of the plant materials and, consequently, the cycling of the nutrients in the soil (Moreira and Siqueira, 2002). The mobilization of nutrients in the soil biomass itself leads to the rapid enrichment of soil nutritional reserves due to the short life of the microorganisms. Thus, in areas with residues that stimulates the increase of microbial biomass, higher are the rates of mineralization of the nutrients that compound the phytomass (Carneiro et al., 2008).

The sorghum residues provided greater accumulation of fresh and dry matter in crambe plants (Table 2). Possibly the answer is due to the physical barrier formed by sorghum straw on the soil surface, avoiding the excessive loss of water by evaporation, considering that the precipitation during the period that comprised the experiment was practically null and the average of relative humidity remained around 50% (Figure 1).

Residues from grasses present high potential for systems in which the presence of soil cover is desired. This is due to the slow decomposition of plant material (Mingotte et al., 2014), which also contributes to the control of weeds, through the physical suppression of development (Favero et al., 2001) and nutrient mineralization, such as Nitrogen, for periods up to 210 days after straw formation (Torres et al., 2005). The mulch management also promotes maintenance of soil moisture, improving water use efficiency, increasing crop productivity and greater weight gain of cabbage head (Carvalho et al., 2011).

Although the treatments containing residues of legume plants and spontaneous plants did not present the highest masses of dry or fresh matter but they also increased the accumulation of dry matter in relation to the fresh matter in crambe plants (Table 2). Again, this is due to the greater speed of decomposition of these species residues, making available the nutrients present in the tissues through the decomposition carried out by the edaphic microorganisms and later releasing these to the soil (Moreira and Siqueira, 2002).

The productive parameters were not affected by the cover crops residues. Average values of 236.47 g L^-1 were obtained for the hectoliter weight of harvested grains, in addition to a mass of 41.77 g per plant and an estimated yield of 960.69 kg ha^-1 (Table 3).

Table 3

Despite the lack of statistical significance between the productivity, the estimated gross revenue for all treatments containing cover crops residues was higher than that obtained in the control treatment (Figure 2). This superiority reflected an increase of approximately 29.16%, 16.66%, 6.38%, 16.32% and 20.64% of the treatments containing residues of pigeon peas, porcine beans, millet, crotalaria and sorghum, respectively, on the gross revenue obtained with the cultivation of crambe on residual of spontaneous plants (US$ 101.62). Thus, although the treatments were not reflected in statistically significant differences for the variables related to the production, the planting of crambe in areas containing cover crops residues increases the monetary gains in addition to the benefits provided to the soil. In agreement with the results obtained by Carneiro et al. (2008), soil management with spontaneous plants is not recommended.

Figure 2

The maintenance of cover crops residues (spontaneous plants, pigeon peas, porcine bean, millet, crotalaria and sorghum) benefits the vegetative development of the crambe crop.

The cultivation of the crambe with cover crops residues does not statistically alter the yield of grains. However, this technique results in economic benefit, with the increase in gross revenue. The use of cover crops during fallow is recommended to the detriment of the management of spontaneous plants.