Physiological quality of soybean seeds stored after industrial treatments with different chemicals

This study aimed to relate the smallest set of variables that compose the quality of soybean seeds lot under study, as well as to evaluate the influence of four industrial seed treatments and their respective slurry volumes on the physiological potential. The experiment was carried out in a completely randomized block design with 4 replicates and 24 treatments. The chemical treatments were: control (T1), micronutrient + polymer + drying powder (T2), bioregulator + polymer + drying powder (T3) and micronutrient + bioregulator + polymer + drying powder (T4). The seeds were stored for periods of 0, 15, 30, 45, 60 and 90 days and were subsequently evaluated for their physiological potential. In each storage period, the variable germination, first count, accelerated aging, emergence speed index, final emergence in the sand substrate, aerial part length, root length and total seedling length were evaluated. The main


Introduction
Ensuring the quality of a seed lot is essential to obtain high productivity since it provides producers with security regarding the physical, sanitary, genetic, and physiological attributes of the seeds. In the field, however, the seeds are exposed to numerous phytosanitary threats, demanding, therefore, the use of chemically treated seeds, to guarantee the maximum physiological potential in the field .
In soybean culture, although widely used in on-farm treatment, the use of cobalt and molybdenum in industrial seed treatment (IST) is still incipiently employed. Both micronutrients are indispensable for success in the biological fixation of atmospheric nitrogen, a process carried out by bacteria of the genus Bradyrhizobium spp., when in symbiosis with soybean culture (Sfredo & Oliveira, 2010).
Besides the addition of micronutrients and biostimulant growth products have also been added to soybean seed treatment mixes, with a report of beneficial effect on plant physiology and the increase in the number of flowers and roots (Klahold et al. 2006). However, the high volume of slurry used in the IST has been reported to be able to accentuate the deleterious effects of the seed deterioration process, especially during storage .
The physiological potential of the seeds can be defined through laboratory analysis after sowing or during storage. In this context, routine analyzes and vigor tests make it possible to compare the quality of seeds, to determine the potential of the seeds submitted to chemical treatments and storage conditions (Marcos Filho, 2015a;Pereira et al. 2020), as well as to identify the lots with superior performance during the storage period and in the field (Guedes et al. 2009).
However, it is important to emphasize that the observation of variables with high correlation must reproduce the quality of the lot and consistently identify the maximum potential of the plants in the field (Johnson and Wichern, 2007). In this sense, the principal component analysis proposes the overlap elimination, identifying the relationship between the information extracted from the data. Thus, there is a need for studies focused on understanding the interrelationship of variables used in tests applied in seed analysis.
Within this context, the hypothesis established is that the physiological quality of seed lots, established by laboratory tests, can be explained by variables with high and low variability among themselves. Given the above, the objective of this study was to identify and relate the most relevant set of tests for the analysis of the physiological quality of treated and stored soybean seeds, depending on the composition of the slurry and storage periods.

Methodology
This quantitative research is characterized by the use of quantification, both in the collection as in the treatment of information, using statistical techniques (Richardson, 1999;Pereira et al. 2018 For each treatment, 2.5 kg of seeds of the soybean cultivar BMX Alvo RR was used. The IST was performed in a continuous seed coating device, which was subsequently placed in kraft paper bags and maintained in laboratory environmental conditions. The test was conducted in a completely randomized block design, with 4 replicates and 24 treatments, so that the seeds were subjected to combinations of the following products: micronutrient (M) (cobalt and molybdenum -CoMo Platinum®, dose: 200 mL 100 kg -1 seeds), bioregulator (B) (kinetine, gibberellic acid, and 4-indole-3ilbutyric Acid -stimulate®, dose: 500 mL 100 kg -1 seeds), polymer (POL) (Disco Ag Green, dose: 100 mL 100 kg -1 of seeds) and drying powder (D) (Fluids, dose: 150 and 400 g 100 kg -1 of seeds).
For each treatment, evaluations were performed before and after the periods of 0, 15, 30, 45, 60, and 90 days. Regarding the treatment and storage of seeds, they were submitted to two different conditions, so that part of the seeds were treated and later stored, and after this, tests and evaluations were installed. The rest of the seeds were stored and treated on the day of test installation. The evaluation of the physiological potential of the seeds was performed for each of the storage periods.
The data obtained were tested for normality through the Shapiro-Wilk test. To verify significant differences among treatments and among storage periods, the Scott-Knott test was used. The t-Student test was applied to verify significant differences between the treatment periods, and for the analysis of the main components, the relationship structure of the variables was identified with the various combinations of IST. The significance level of 5% was considered in all tests. Data were analyzed using software R, version 4.0.2 (R Core Team, 2020).

Results and Discussion
The analysis of variance indicated a significant difference (p ≤ 0.05) between the evaluated treatments and storage periods. In Table 1, the variables show significant differences between before and after storage (p < 0.05), except for the variables AL and AA.   The treatment of soybean seeds performed before sowing with different combinations of chemicals and slurry volumes can provide a phytotoxic effect on seedlings (Brzezinski et al. 2015). Despite the decrease in germination, seed storage for up to 60 days (Table 2) did not compromise the potential for seed commercialization, since in SGT the average values of normal seedlings are above 80%, a level established as a minimum guarantee for the commercialization of seed lots in Brazil (Brazil, 2013). In this context, Smaniotto et al. (2014) suggested that the humidity resulting from the IST favored deterioration, especially during storage. In this same perspective, Zorato & Henning, (2001) pointed out that after 90 days of storage, loss of seed quality was inevitable. In Table 3, it was found that treatment T1 showed superior results about the others, for all response variables, except for ES, in which treatments T1 and T2 did not differ statistically from each other, however, they surpassed the others. In this regard, Taylor & Salanenka, (2012) pointed out that, unlike paper tests, in the emergency sand test (ES), the components of the slurries were diluted in the substrate, thus increasing the sensitivity of the seeds, especially in high vigor seeds. Associated with seed treatment, various products such as biostimulants, micronutrients, and polymers provide increased productivity (Klahold et al. 2006). However, Brzezinski et al. (2017) observed that the high volume of slurry, associated with long periods of storage, harmed seed quality. In the case of soybeans, the presence of a micropyle favored the permeability of the membranes, so that during germination, the chemical in contact with the root caused phytotoxicity and compromised the development of the roots (Peske & Peske, 2011).
Regarding the correlation analysis among the variables responses, it was found that all are strong and positive, showing that these variables have good properties to be part of the analysis of main components. Table 4 shows the percentage of the total variability explained by each factor and the accumulated percentage. When observing the first factor, it corresponds to about 87.43% of the total variance, with the first two factors reaching about 92.36% of the total variance; in these circumstances, main component 1 was responsible for 87.43% and main component 2 was responsible for 4.92% of the data variations. Thus, if two factors are selected, the reduction in size from eight original variables to two factors is considered satisfactory. Therefore, only the first two factors were used for the composition of equations (1) and (2). (1) (2) (CC BY 4.0) | ISSN 2525-3409 | DOI: http://dx.doi.org/10.33448/rsd-v10i2.12279 6 In equation (1), referring to the first factor, the variables ES and SL were highlighted, being called final emergency factor in sand and total length of seedling as the factors that most report to each other, confirming the results observed by Vanzolini et al. (2007), in which lots with a high percentage of normal seedlings tend to present satisfactory performance regarding the emergence in the sand. In this respect, Brzezinski et al. (2017) suggested that seeds from high vigor lots had the same tendency to present a high percentage of ES and SL. On the other hand, these factors show that they are negatively correlated with the others, especially with RL and AL, which presented non-expressive relationships and marked discrepancy with the other variables. Regarding equation (2), referring to the second factor, the variables SGT and AA were highlighted, which can be called the contrast factor of the Standard Germination and Accelerated Aging Test (Table 5). Such results suggest that these variables explain most of the total variation, conferring the existence of a high correlation between SGT and AA. In this same perspective, Schuab et al. (2006), when assessing the physiological quality of soybean genotypes, observed high correlation values between the AA, SGT, and FC test. On the other hand, the results obtained for ESI and AA indicated a negative correlation among different soybean genotypes studied.
To determine the number of retained factors, it was found that as the first two factors generated from this analysis have eigenvalues > 1 ( > 1) (Kaiser, 1958;Fraga et al. 2015) and were responsible for 92.36 % of the total variance in the data set, the first two factors were retained. To verify the importance of each variable in the construction of the two factors, the correlations between the original variables and the main components and their weighting coefficients were calculated (Table   5).
In this sense, it is important to highlight that, during the aging process, the germinative capacity and the physiological potential of the seeds are affected, so that it starts with the production of free radicals and lipid peroxidation and culminates in disorders of the membrane and tissue death (Marcos Filho, 2015b). Considering equations (1) and (2), there are highly expressive relationships of the seed lot, mainly on the influence of ES and SL by Factor 1. In this perspective, it is known that the analysis of main components helps to understand the information about the studied variables, so that the least possible loss of information from the database under study is sought.

Conclusion
The increase in the volume of slurry reduced the physiological quality of the seeds during all storage periods. It is important to highlight that the seed treatment directly affects the physiological potential, so that the vigour and viability of the seeds become compromised with the increase of products added to the slurry, especially when it comes to repeating the seeds stored after the industrial treatment. The analysis of the main components allowed to reduce the number of variables, in order to the SGT and FC components were the ones that most explained the total variability of the original data. As future work, it is necessary to evaluate different concentrations of essential oil of basil in order to verify its allelopathic effect, that is, inhibitory on the physiological quality of seeds of sedge weed (Cyperus rotundus).