Introduction

Radish (Raphanus sativus L.) is a vegetable native from the Mediterranean region (^{Rodrigues et al., 2013}), belongs to the Brassicaceae family, the same family of cauliflower, oilseed radish, rocket, cabbage, and others. The plants usually have small sizes, tuberous roots, and green compound leaves with leaflets and lobes (^{Guimarães and Feitosa, 2014}).

The consumption of vegetables has increased in the last years due to the search for healthier foods with nutraceutical properties (^{Cunha et al., 2017}), considering that the population is aware of the benefits and properties of these foods. It increases the responsibility and challenges for rural producers who need to increase crop yields and search for tools to control and mitigate the occurrence of pests and diseases and, also, search for compatible sustainable production systems that ensure the availability of foods to consumers and, consequently, the growth of the food production sector (^{Montezano and Peil, 2006}).

The world radish production is estimated at 7 million Mg year^-1. Japan is one of the largest radish producers (^{Linhares et al., 2010}; ^{Ito and Horie, 2008}). In Brazil, radish is grown mainly in the South and Southwest regions, showing yields from 11 to 30Mg ha^-1 and a production cost of BRL (R$) 10,000 to 14,000 per hectare (^{Melo, 2017};^{Minami, 1998}). Despite present in several Brazilian recipes, radish crops have not been emphasized in Brazil regarding the crop area and production volume (^{Cecílio Filho and May, 2002}).

Radish presents the lowest expressiveness in Brazil among marketed vegetables (^{Puliti et al., 2009}). Therefore, it can be grown in small size properties along with green belts of large cities, enabling a faster financial return and higher income in periods between seasons of crops of longer cycles (^{Cardoso and Hiraki, 2001}; ^{Oliveira et al., 2010}). Thus, radish is a good alternative crop, mainly for small farmers, since it can be grown throughout the year and between other crop seasons (^{Bonela et al., 2017}).

The crop spacing between rows and between plants and the number of plants per pit are factors that define the plant population to be grown. Climate, soil, and plant chemical and physical characteristics, crop management, and cultural practices should be considered to determine these factors (^{Severino et al., 2006}). Studies on radish crops for the Cocais administrative region of the state of Maranhão, Brazil, are scarce. In this context, this study was conducted aiming to evaluate the agronomic performance of radish (R. sativus L.) cultivars grown under different spacings.

MATERIAL AND METHODS

The experiment was conducted with radish crops at the Experimental Field of the Federal Institute of Education, Science, and Technology of Maranhão, in Codó, in the mesoregion East of the state of Maranhão, Brazil (4°26'51''S, 43°52'57''W, and altitude of 48m) (^{Castro Júnior et al., 2015}).

According to ^{Köppen and Geiger (1928)}, the climate of the region is Aw; the mean air temperature is 27.4ºC; the mean annual rainfall depth is 1.526mm. August is the driest month, with a rainfall depth of 12mm, and March presents the highest mean rainfall depth (307mm). September is the hotter month with a mean temperature of 28.8°C. The soil of the area was classified as Typic Quartzipsamment (^{Santos et al., 2018}).

Soil samples from the 0.0 - 0.20m layer were collected and sent to the Laboratory of Soil Analysis of the Federal University of Piauí (UFPI) for soil chemical characterization; the results are shown in Table 1.

Soil chemical analysis indicated that liming was not needed. A soil organic fertilizer (cured caprine manure) was used at 20L per plot of 1m². A chemical fertilizer (N-P-K) was applied according to the soil analyses and recommendations of the Manual of Fertilization and Liming (^{Aquino et al., 1993}).

Urea (CH₄N₂O), simple superphosphate (P₂O₅), and potassium chloride (KCl) were applied to the radish beds. The soil chemical fertilizers were applied in two phases: soil fertilization at planting, using P and K; and a topdressing at 15 days after planting, using N and K.

The area was chosen and delimited, and a brush cutting and weeding were carried out through a harrowing of the area. The beds were then manually raised using a hoe. Cured caprine manure was applied to the beds and incorporated into the soil, and the area was leveled using a rake. The plots were then established by dividing beds of 2 × 1m into two plots.

The following cultural practices were carried out during the crop cycle: adding soil to the plants' base, removal of weeds, manual hoeing, and plant health protection. The plants were harvested at 30 days after sowing, on July 18, 2017. A localized irrigation method was used, with a micro-sprinkler system.

The evaluated variables were: total fresh weight (g plant^-1), using a digital balance; root fresh weight (g plant^-1), for which leaves were cut at their base to remove them from the roots, and the roots were weighed in a digital balance; root-mean diameter (mm), using a caliper to measure the transversal diameter; root mean length (mm), measuring the root length of each plant with a caliper, and dividing the result by the number of plants in the evaluation area; root yield (kg ha^-1), which is the weight of radish roots produced divided by the planted area.

A randomized block experimental design was used, in a 2×2 factorial arrangement, with five replications. The treatments consisted of two cultivars (Sparkler Ponta Branca and Saxa) grown under different spacings between plants (5 and 8cm), totaling 20 experimental plots, with 6 evaluated plants per plot.

Thus, the treatments were: Sparkler Ponta Branca with a spacing of 5cm (T1); Sparkler Ponta Branca with a spacing of 8cm (T2); Saxa with a spacing of 5cm (T3); and Saxa with a spacing of 8cm (T4).

The data found were subjected to analysis of variance (ANOVA) by the F test. The means were evaluated through comparison by the Tukey's test at 5% probability, level using the Assistat^® 7.7 beta program (^{Silva and Azevedo, 2016}).

RESULTS AND DISCUSSION

According to the analysis of variance (Table 2), the factor cultivar (Sparkler Ponta and Saxa) had no significant effect (P ≥ 0.05) on the total fresh weight (TFW), root fresh weight of (RFW), root mean diameter (RMD), root mean length (RML) and root yield (RY) of radish plants; and the factor spacing between plants (5 and 8cm) had no significant effect (P ≥0.05) on the variables TFW, RFW, RML, and RMD. However, the spacing between plants (5 and 8cm) had a significant effect (P <0.01) on RY, showing lower RY when using the spacing between plants of 8cm.

These results are explained by the effect of climate conditions since oscillations in soil moisture and temperature throughout the crop cycle affect the yield and quality of radish roots (^{Costa et al., 2006}).

The cultivar (Sparkler Ponta and Saxa) and spacing between plants (5 and 8cm) had no significant effect (P ≥ 0.05) on TFW (Table 3). ^{Caron et al. (2004}) found similar results that corroborate those found in the present work; they explain that the shoot fresh weight of radish plants determines the leaf area and is directly correlated to the production and distribution of photoassimilates and allocation of biomass, mainly in leafy vegetables.

The cultivar (Sparkler Ponta and Saxa) and spacing between plants (5 and 8cm) had no significant effect (P ≥ 0.05) on RFW (Table 4). These are similar results to those found by ^{El-Desuki et al. (2005}), ^{Rezende et al. (2006}) and ^{Bonela et al. (2017}), who found no significant differences for RFW in radish plants of the cultivar Crimson Giant, Saxa, Sparkler Ponta Branca, Red Jewel F1, and No. 25.

The cultivar (Sparkler Ponta and Saxa) and spacing between plants (5 and 8cm) had no significant effect (p≥0.05) on RMD (Table 5). Despite the lack of significance between the treatments, the presence of radish roots with problems, such as sponge-like texture, was not found. These problems result in radishes with a spongy texture, insipid taste, and cracks, thus compromising their quality (^{Filgueira, 2003}).

^{Caetano et al. (2015}) evaluated radish crops and found RMD of 19.47 and 18.20mm, which were similar results to those found in the present study. However, ^{Salgado et al. (2006}) and ^{Vitti et al. (2007)} found RMD of 3.5cm for radish plants grown under organic soil fertilization.

The cultivar (Sparkler Ponta and Saxa) and spacing between plants (5 and 8cm) had no significant effect (p≥0.05) on RML (Table 6). ^{Cortez et al. (2010)} reported that tall plants are not a desirable characteristic to increase radish yields.

The cultivar (Sparkler Ponta and Saxa) had no significant effect (p≥0.05) on RY. However, the spacing between plants had a significant effect (p<0.01) on RY, showing lower RY when using the spacing between plants of 8 cm (Table 7). According to ^{Cecílio Filho et al. (1998}), mineral nutrition for plants has a high effect on the crop, affecting root qualitative aspects and, mainly, the crop yield.

^{Batista et al. (2013)} found similar results for radish crop yield, with the highest total root yield of 11.81g plant^-1, which was found at 21 days after the incorporation of 15.6Mg ha^-1 of Calotropis procera biomass.

CONCLUSION

The spacing between plants has no significant effect on the yield of Sparkler Ponta Branca radish cultivar but significantly affected the yield of the cultivar Saxa, where higher yields are recorded when the spacing between plants was of 5 cm.