Nitrogen doses and splitting in top dressing in the production, quality, and macronutrients content in zucchini (Cucurbita pepo) seeds

Despite the importance of the nutrients, mainly nitrogen, in the formation and chemical composition of the seeds, little research has been done to verify the relationship between the nutrients supplied to the plants and the quality of the seeds produced. The objective of this study was to evaluate the influence of nitrogen doses and splitting on yield, quality and macronutrients content in zucchini seeds. Two experiments (crop 2014 and crop 2015) were installed, with thirteen treatments in the factorial scheme 4x3+1, with four nitrogen doses in top dressing (62.5; 125.0; 187.5; 250.0 kg ha), three splitting (1/6+1/3+1/2; 1/4+1/2+1/4; 1/3+1/3+1/3 of total dose in each split), and one treatment without N fertilization in top dressing (dose 0), with four replications. Seed production and quality and macronutrient content in seeds were evaluated. There wasn’t significant effect on number of fruits and seeds produced per plant. However, in crop 2014 the higher the doses of N in top dressing, the higher the seed weight per fruit and per plant, the weight of 100 seeds and the vigor (first count in germination test). In crop 2015, seeds with the best quality (germination and first counting) were obtained in the 1/4+1/2+1/4 splitting. The decreasing order of the macronutrient content in the seeds was N>P>K>Mg>S>Ca.


Introduction
A balanced fertilization is essential in the production of seeds, due to the fact that the nutrients participate in the seed formation, development and maturation phases, being part of the membrane constitution and in the accumulation of lipids, carbohydrates and proteins, reflecting in the size, weight and chemical composition of seeds (Carvalho & Nakagawa 2012).
However, despite the importance of the nutrients in the formation and chemical composition of the seeds, little research has been done to verify the relationship between the nutrients supplied to the plants and the quality of the seeds produced (Cardoso 2011;Magro et al. 2010).
In zucchini, there are fertilization recommendations for the production of commercial immature fruits (Araújo et al. 2013;2015). However, when the focus of the crop is the production of seeds, this fertilization may be insufficient, as the plants stay for longer time in the field and the seeds, still in the beginning of formation in the immature fruit, are the main drains in the ripe fruit (Cardoso 2011).
In the reproductive phase, the nutritional requirement of the plants is more intense and nutrients are intensely translocated to seed formation. Among these nutrients, nitrogen is prominent in seed formation, as it is involved in the constitution of proteins, starch synthesis, embryo formation and reserve tissue. Nitrogen can also influence the physiological quality of the seeds, but their effects vary with the environmental conditions and the stage of development of the plant in which the fertilizer application occurs (Carvalho & Nakagawa 2012).
Nitrogen is the second most extracted nutrient by immature fruits in zucchini (Araújo et al. 2015), with part being supplied in planting fertilization, and the rest distributed in three equal applications in top dressing fertilization (Trani & Raiji 1997), because it is a nutrient easily lost by volatilization and leaching and the plants require different amounts during the developmental stages (Cardoso 2011;Corrêa et al. 2014;Cardoso et al 2016). But there are few researches about top dressing fertilization in vegetable seed production and quality.
The objective of this study was to evaluate the influence of nitrogen doses and splitting on yield, quality and macronutrients content in zucchini seeds.

Methodology
Two experiments were carried out, one in 2014 (Crop 2014) and another in 2015 (Crop 2015), both in the São Manuel Experimental Farm, located in the municipality of São Manuel, Brazil, belonging to the School of Agriculture (FCA) of the Sao Paulo State University (UNESP). The geographical coordinates are: 22° 46 'South latitude, 48° 34' West longitude and 740m altitude. The climate of the São Manuel region is Cfa, warm temperate (mesothermal) and humid (Cunha & Martins 2009) and the soil is a Typical Red Dystrophic Latosol. It is a sandy soil with 84% sand.
In both crops, the following characteristics were evaluated: number of ripe fruits per plant, average fruit weight, length of the fruit, seed production (number and weight) per fruit and per plant, germination and vigor of seeds.
For the germination and vigor test (first count), the seeds were placed on two sheets of pre-moistened towel paper, covered by another sheet, and placed vertically in a germination chamber type BOD at a temperature of 25ºC (Brasil 2009), with four replicates of 50 seeds. First count (only radicle protrusion) was evaluated at the 4 th days after sowing (DAS), and total germination (normal seedlings) was performed at 8 th DAS, both expressed in %.
Research, Society and Development, v. 10, n. 10, e328101018708, 2021 (CC BY 4.0) | ISSN 2525-3409 | DOI: http://dx.doi.org/10.33448/rsd-v10i10.18708 4 The determination of the macronutrients (N, P, K, Ca, Mg, S) content in seeds was performed only in crop 2015. For this, on the day of the seed extraction, a seed sample per plot was packed in a paper bag and placed in a forced air circulation oven at 65 °C until reaching a constant weight (± 48 hours). After drying, the samples were analyzed according to methodology presented by Malavolta et al. (1997), and results were expressed in g kg -1 of dry weight (DW).
Data were submitted to variance and regression analysis to verify the effect of nitrogen doses, defining the best fit according to combination of significance and higher coefficient of determination, and Tukey's test (p> 0.05) for determination of the effect of splitting. The data were processed by the Sisvar 5.3 computer statistical analysis system (Ferreira 2011).

Results
To crop 2014, the treatments did not influence the number of fruits (1.5 fruits per plant), average fruit weight (1.1 kg) and fruit length (31.7 cm), also did not influence some characteristics related to the seed production as the number of seeds per fruit (273 seeds) and number of seeds per plant (379 seeds).
For the characteristics seed weight per fruit, first count and final germination, there was interaction between the nitrogen doses and splitting, and for weight of 100 seeds and seed weight per plant there was an effect only for doses of N. For the 100 seed weight, the effect of the doses was linear, with an estimated mean increase of 30.1% in the dose 250 kg ha -1 of N compared to the control (dose 0) (Figure 1). Despite there is no difference in seed number per fruit, as the 100 seed weight increased with doses of N, it was observed a linear effect for weight of seeds per fruit, in all the splitting, with an increase of 36.7%, 26.9% and 28.2% in the seed production (g) per fruit in the dose 250 kg ha -1 of N compared to the control (dose 0) in the 1/6+1/3+1/2, 1/4+1/2+1/4 and 1/3+1/3+1/3 splitting, respectively (Figure 2A). For seed production (g) per plant, the means were adjusted to the linear model, without interactions, with an increase of approximately 34.7% in the seed weight at the dose 250 kg ha -1 of N when compared to the control, regardless the splitting ( Figure 2B). The averages of the first seed germination count in the 1/6+1/3+1/2 and 1/4+1/2+1/4 splitting increased the higher the doses of N, with a maximum of 90.2 % and 87.9% in the highest dose, respectively. In the 1/3+1/3+1/3 split, there were not effect for nitrogen fertilization, with a mean of 78.7% at the first count ( Figure 3A). The seed germination averages in the 1/6+1/3+1/2 splitting had a linear increase ranging from 70.8% (control) to 87.5% (highest dose). In the 1/4+1/2+1/4 and 1/3+1/3+1/3 splitting, the quadratic effect was obtained, with an increase in germination percentage with a maximum estimated in 87.3% at dose 163 kg ha -1 of N and 83.4% in the dose 115 kg ha -1 of N, respectively ( Figure 3B). For the first count and the total germination there was difference for splitting, doses and interaction between factors.
The total germination averages in the 1/6+1/3+1/2 splitting did not respond to the increase of N doses, with an average of 67.5%. In the 1/4+1/2+1/4 splitting, the germination averages responded with a cubic effect, with maximum germination estimated in 100% at the dose 185 kg ha -1 of N. Behavior similar was observed in the 1/3+1/3+1/3 splitting with maximum germination estimated in 100% at the dose 106 kg ha -1 of N ( Figure 4B). Research, Society andDevelopment, v. 10, n. 10, e328101018708, 2021 (CC BY 4.0) | ISSN 2525-3409 | DOI: http://dx.doi.org/10.33448/rsd-v10i10.18708 7 The factor splitting and the interactions between doses of N and splitting were not significant for all nutrients content in the seeds. There was a significant effect of doses of N on the contents of N, P, K and Ca. For Mg and S contents, there was no effect of the treatments, with means of 5.77 and 1.78 g kg -1 DM, respectively.
The nitrogen content in the seeds increased the higher the doses of N, ranging from 51,6 to 59,1 g kg -1 DM, an increase of 15.4% in the highest dose (250 kg ha -1 ) when compared to the control (dose 0) ( Figure 5A). The phosphorus content was adjusted to the quadratic equation, with maximum values estimated for P content of 10.1 g kg -1 DM at the dose 197 kg ha -1 of N ( Figure 5B). For potassium and calcium contents it was also observed quadratic equation, with minimum values estimated for K content of 8.3 g kg -1 DM at the dose of 158 kg ha -1 of N and for Ca content of 2.9 g kg -1 DM at the dose of 166 kg ha -1 of N ( Figures 5C and 5D).

Discussion
The fact that no differences were observed for number of fruits per plant in crop 2014 could be due to the fruits being harvested ripe, that is, when it is normal to get only 1 or 2 fruits per plant, because the other fruits are aborted. But, when fruits are harvested before ripening, it is common to harvest more than ten fruits per plant, because, in this case, the fruits are weak drains, while ripe fruits are strong drains that promote abortion of young fruits in the plant (Stephenson et al. 1988;Cardoso & Souza Neto 2016).
For the characteristics related to seed production, number of seeds per fruit is dependent of number of visits of bee pollinators (Stephenson et al. 1988) that, probably, was about the same for all flowers.
The application of nitrogen in top dressing aims to complement the nutrition providing this nutrient during the development of the seeds, period of greatest nutrients demand, as illustrated in Figure 1.
In relation to Figure 2, the greater production of seeds in weight was due to the greater 100 seed weight, because there was not an increase in number of seeds produced per fruit and per plant. Nitrogen, together with potassium, are usually the most extracted by plants and N is usually the most accumulated in seeds (Cardoso 2011), because they participate in the constitution of proteins, starch synthesis, embryo formation and it is important in the initial development of the embryo during germination (Kolchinski & Schuch 2004). The increasing in seed production was linear, until dose 250 kg ha -1 of N, that is 200% higher than the dose recommended by Trani and Raij (1997) for zucchini (immature fruit) production, showing that, probably, the need for N for seed production should be higher than the need for immature fruit production.
The increase of the N dose when applied in higher quantities in the flowering and fruiting phases allowed an increase in the viability and vigor of the seeds (Figure 3). According to Carvalho and Nakagawa (2012), nitrogen can influence the physiological quality of the seeds, but their effects vary with the environmental conditions and the stage of development of the plant in which the fertilizer application occurs.
About the first count and the total germination (Table 1), nitrogen fertilization done in the highest proportion in the 2 nd application, in the 1/4+1/2+1/4 splitting, may have favored the physiological quality because it provides a higher amount of N, compared to the other two splitting, just in the phase of the cycle that demands the greatest quantity of this nutrient which is the beginning of the development of fruits and seeds. Nitrogen is the nutrient that has the greatest influence on the vigor of the seeds and when applied in the beginning of flowering, the same is directed to the seeds (Cardoso 2011;Carvalho & Nakagawa 2012).
The germination values obtained ( Figure 4) were smaller than those of the first count because of the evaluation method adopted, that is, for the first count, the radicle protrusion of at least 4 mm in length was considered and for final germination only normal seedlings, according to Brasil (2009).
As the results of this research ( Figure 5A), Toledo et al. (2009) also observed a linear increase in N content in bean seeds with increasing N dose applied at the flowering and grain filling stage. This increase in content of N is due the greater availability of N in soil to the plants and because N is easily redistributed from the leaves to the fruits and seeds.
In comparison to determination of the macronutrients content in seeds ( Figures 5A-D), Patel (2013), in pumpkin, obtained average contents of 48.37; 12.33; 8.09; 0.46 and 5.92 g kg -1 DM of N, P, K, Ca and Mg, respectively, while El-Adawy & Taha (2014), in pumpkin seeds, related 30.7 g kg -1 of N, 10.9 g kg -1 of P, 9.8 g kg -1 of K, 1,3 g kg -1 of Ca and 4.8 g kg -1 of Mg. The values are a little different probably due differences in cultivar, soils, fertilization, and climatic conditions. But the decreasing order of the macronutrients obtained in the seeds are the same: N>P>K>Mg>S>Ca. Just S was not evaluated by the other authors and it is not possible to confirm the order.
Probably, the greatest accumulation of N in the seeds is due to the fact that this nutrient participates in the constitution of the proteins, starch synthesis, formation of the embryo and reserve tissue, viability and vigor, and it is important in the initial development of the embryo during germination (Kolchinski & Schuch 2004;Carvalho & Nakagawa 2012). Phosphorus is important in phytin composition, the main form of energy storage in seeds (Carvalho & Nakagawa 2012;Kano et al. 2012).
These nutrients (N and P) are the most important in seed formation of most species, as reported by several authors (Kano et al. 2011;Cardoso 2011;Cardoso et al. 2016). On the other hand, Ca is the nutrient with lower seed content, a fact probably related to the small mobility in the plant by the phloem (Grangeiro et al. 2007;Malavolta 2006). According to Cardoso et al. (2016), the accumulation of Ca in the seeds should only occur by absorption and transport during the maturation of the seeds, without redistribution from the leaves in senescence to the seeds.

Conclusion
The treatments did not result in an increase in fruit production and number of seeds produced per plant.
In crop 2014, the higher the nitrogen doses, the higher the seed production (weight) per fruit and per plant and the 100 seeds weight.
In the seeds, the decreasing order of the macronutrients content was N>P>K>Mg>S>Ca.
The higher the nitrogen dose, the higher the nitrogen and phosphorus content in the seeds.