J. curcas belongs to Euphorbiaceae family; this species shows large agricultural potential, highlight for seed yield, oil quantity and quality, aiming to biodiesel production (Tiwari et al., 2007; Castillo et al., 2014). The cultivation of J. curcas shows several advantages in relation to oil production: it is perennial, rustic, easy management, reaching 37.49% oil seed content, and high oil quality for biodiesel (Singh et al., 2016). J. curcas may be adequate for intercropping with other species because it is a shrubby plant and associated with wider space between rows the planting of other crops might be feasible (Silva et al., 2012).

The need to supply the demand for foods by means of scarce resources through smallholders is increasingly the adoption of intercropping systems for crop production (Machado, 2009). The intercropping has been used for many smallholders in Brazil, even more in relation to farmers who search higher land use efficiency and greater economic return (Veronesi et al., 2014), besides the generation of viable alternative to increase food offer. Nevertheless, seed yields obtained by  in some locations in the world are limited because of feasible agricultural practice adopted (Liyama et al., 2013).

In relation to J. curcas production traits, oil yield depends on the vegetable features as number of branches,  crown   projection,   plant   height   and  crown volume, and production features; seed yield, seed weight, shell mass and seed oil content (Rao et al., 2008). The management adopted may result in negative or positive influence of crops in intercropping for production of the major crop; the positive effects may be related to improvement of soil chemical and physical properties, and negative ones might be due to possible water, light and nutrients competition (Tjeuw et al., 2015).

To improve positive interaction in intercropping, knowledge on the features of the crops associated is necessary, because without this information the farmers are going to commit many mistakes before achieving higher economic returns. This study was carried out with the aim to assess the biometric and productive features of J. curcas in intercropping with forage grass and grain crops species.

Site description and soil

The experiment was carried out in the district of Itahum, city of Dourados, state of Mato Grosso do Sul, Brazil, at the coordinates 22°05’44" S and 55°18’48" W, enabled by a partnership between Brazilian Agricultural Research Corporation (Embrapa Western Agriculture) and Paraíso Farm. The soil is classified as Typic Haplortox (Santos et al., 2013), with average clay content of 200 g kg^-1. Long-term monthly rainfall averages, as well as actual rainfall recorded during the trial is presented in Figure 1.

J. curcas was sown in November 2006, on Paraíso Farm using a no-till system, by depositing three seeds per hill. After emerging, only the most vigorous seedling was left in each hill. Planting rows were spaced at 3 m and plants were spaced at 2 m within the row. In    2006/2007    and    2007/2008    growing   seasons   the   usual management practices were applied to the field.

Experimental site and design

The treatments were installed in experimental plots comprised of four rows of J. curcas with six plants per row (144 m² per plot), the treatments are shown in Table 1. In order to evaluate plant height, crown diameter, stem diameter, and number of branches, the experimental was laid out in randomized complete block design with four repetitions, in a joint analysis in factorial scheme 11 × 2 (11 crop management system and two growing seasons). To assess 100-seeds weight, seed yield and seed oil content, factorial scheme 11 × 3 (11 crop management system and 3 growing seasons) was performed.

J. curcas fertilizer rate was applied annually with 32 kg N ha^-1, 80 kg P₂O₅ ha^-1 and 80 kg K₂O ha^-1, through the formulation of 08-20-20 (400 kg ha^-1). Fertilizer rates was carried out manually close to the planting row. The fertilizer was divided in two applications for each growing season (2008/2009 and 2009/2010) (50% in October and 50% in January). In addition, 50 kg N ha^-1 using urea as N source was applied in January 2009 and January 2010. Fertilizer rates recommendation followed the suggestion of Laviola and Dias (2008). The treatments with crop rotation (9, 10 and 11) were managed and fertilized according to recommendations for each crop. The remainder of the treatments did not receive any fertilizer rate. Intercropped forage grass and cover crops species were managed by mowing according to management height indicated by the research for each species. The resulted stubble was uniformly distributed on soil within the plot for mulching (Silva et al., 2012).

Variables assessed

In J. curcas trees, stem diameter was determined with the assistance of digital caliber in six plants in each experimental plot; this measurement was accomplished in plant collar. Plant height and crown diameter were determined with graduated ruler in six plants in each experimental plot that was measured from soil surface to the top branch of J. curcas. The crown diameter was measured transversely to the row, at the ends of the largest side branch of the plant. The number of branches from the most vigorous plants were determined by counting the vertical direction (from root to shoot) when the data were recorded and then filled in the final harvest of each growing season. In order to determine J. curcas seed yield, six plants in each experimental plot were harvest manually. Five harvesting time were conducted from December to July in each growing season (2008/2009, 2009/2010, and 2010/2011).

After harvesting, ripe and dried fruits in each experimental plot was stored in paper bags and naturally ventilated until constant weight. After the fruits dried, it was accomplished the threshing and weighted of the dried seed, and determined the seed yield and 100-seeds weight. The analysis of seed oil content in J. curcas grains were accomplished following the method of Soxhlet extraction, according to Lara et al. (1985).

Statistical analysis

The database were submitted to analysis of variance (ANOVA) and in case of significant difference (p<0.05) the means were compared by Tukey test of means with the assistance of the statistical software SISVAR.

Biometric traits of J. curcas in intercropping system with forage grass and grain crops

The crop management systems evaluated in this study did not affect (p>0.05) biometric traits as plant height, crown diameter, stem diameter and number of branches of J. curcas on average of 2008/2009 and 2009/2010 growing seasons (Table 2). These results guide to an opportunity to integrate J. curcas in intercropping with grass and crops. Intraspecific and interspecific plant competition in intercropping is a challenge to be overcome to implement a profitable production system. The natural increment of plant height form 2008/2009 to 2009/2010 growing season was 19%, which was predictable since J. curcas reached its adult height in the fourth year after planting; this way in three years after planting the plants were in vegetable development. In the first growing season, J. curcas height did not differ among the crop managements evaluated, which may be due to the initial development of  the grain crops and forage grass species in intercropping with J. curcas. In 2009/2010, plant height in crop rotation system-1 differed from Brachiaria ruziziensis + Stylosanthes species, but in relation to the other treatments, no significant difference in plant height was observed (Table 2). The increase of J. curcas height indicated that these species evaluated in intercropping may not compete hardly for natural resources with J. curcas that can compromise its development in height. Nevertheless, interspecific competition in inter-cropping of Jatropha with crops has already been mentioned by Tjeuw et al. (2015), who found negative effects on Jatropha height due to moisture and nutrient competition with maize.

J. curcas intercropping system and monocrop did not affect the crown diameter of J. curcas on average of 2008/2009 and 2009/2010 growing seasons (Table 3). With exception of the intercropping of J. curcas with B. ruziziensis that remained without alterations, the other treatments showed higher crown diameter in the growing season 2009/2010 (Table 3). 

It is possible to infer that the spread of roots in deeper layers for J. curcas avoids higher competition for natural resources from soil. As reported by Sánchez et al. (2003), the establishment of roots from grain crops or forage grass species in surface layers and the trees in deeper layers decrease the competition in soil for water and nutrients. The increasing in crown diameter from 2008/2009 to 2009/2010 growing season was 16%, which is quite important due to the energy that the plant needs for growth that comes from the photosynthesis, this way, size of the crown diameter is related to the capacity of assimilate carbon and turn into energy (Larcher, 2004). Thus, the size of crown diameter is associated with the photosynthesize capacity, which is expected that higher crown diameter in relation to higher assimilation of CO₂ may result in increasing grain yield for J. curcas.

The crop management system evaluated showed significant difference only in 2008/2009 growing season, which showed depletion of the crop rotation system-2 on the stem diameter in comparison to intercropping of J. curcas with Chrysolopus spectabilis, the other treatments remained without alteration on stem diameter (Table 4). Nevertheless, this negative effect on stem diameter promoted by crop rotation system-2 was not confirmed in the following growing season (Table 4). In 2009/2010 growing season was observed increment of 19% of stem diameter in comparison to preceding growing season. However no significant difference was observed among the crop management system on average of the two growing seasons (Table 4). Based on stem diameter average of the two growing seasons, these results did not confirm competition for intercropping J. curcas with forages grass or grain crops species evaluated in relation to stem diameter of J. curcas. The stem diameter is positively correlated to root development (Fakuta and Ojiekpon, 2009), thus, plants with higher stem diameter is expected to have better nutrition and higher tolerance to drought stress due to higher root volume to explore the soil for water and nutrients. The absence of significant difference of crop management system in stem diameter of J. curcas pointed out a possibility for further researches to investigate the root development of vegetable species integrated in the production system to assure the absence of root competition through time of plant growth.

Numbers of branches were higher in 2008/2009 growing season for the treatments; intercropping J. curcas with Stylosanthes spp., B. ruziziensis, Brachiaria humidicola, Panicum maximum cv. Massai, Cajanus cajan and crop rotation system-3 (Table 5). Nevertheless, in the following growing season (2009/2010), the crop management system did not affect number of branches (Table 5). The number of branches on average of two growing seasons was not affected by the crop management system evaluated (Table 5).  

The species in intercropping with J. curcas did not affect its vegetable development. J. curcas reach higher seed yield in intercropping with crop rotation system-2 (maize off-season/Crambe abyssinica/soybean/peanut) and crop rotation system-3 (cowpea/radish/maize/cowpea) in comparison to the other species evaluated in intercropping. The J. curcas seed yield is lower in intercropping with forage grass species due to interspecific competition.

The leguminosae Cajanus cajan cv. Anão and Crotalaria spectabilis showed intermediary result in terms of seed yield, which was attribute to lower interspecific competition with J. curcas and maybe biologic nitrogen fixation available for J. curcas.

The authors are grateful to Agricultural Research Corporation (Embrapa Western Agriculture), Foundation to Support the Development of Education, Science and Technology of the State of Mato Grosso do Sul (Fundect-MS), Paraiso Farm, Petrobras, and National Council for Scientific and Technological Development (CNPq) for financial support and scholarship.

The authors have not declared any conflict of interests.