MATERIALS AND METHODS

The experiment was conducted in a greenhouse belonging to the Department of Crop Science of the University of São Paulo, 'Luiz de Queiroz' College of Agriculture, Piracicaba, State of São Paulo (SP), Brazil (22°42'32.0" S, 47°37'43.1" W), from November 2016 to February 2017.

The herbicide metsulfuron (Accurate®, 600 g kg^-1, FMC Química do Brasil Ltda., Brazil) was applied at a rate of 2.4 g of active ingredient (a.i.) ha^-1, before sowing soybean. The maximum recommended rate for weed control in winter management (off-season) (^{Rodrigues and Almeida, 2018}).

The experiment was conducted in a 2×5 completely randomized factorial design, with four replications. The factors were composed of two soybean cultivars (STS and non-STS) and for five periods between application and sowing of soybean (0, 15, 30, 45, and 60 days). Soybean cultivars BMX Garra RR2/STS and M 6410 IPRO (non-STS) were used. The safety interval between application and sowing of soybeans is 60 days (^{Rodrigues and Almeida, 2018}), which is why the periods chosen were a maximum of 60 days. Therefore, the 60 days are considered the control since it is safe for STS or non-STS soybean.

The experimental units were 5 L pots filled with medium texture soil (Table 1). Before the onset of the experiment, the two cultivars were subjected to a preliminary emergence test, with results higher than 90% at 15 days after sowing (DAS), for both.

Table 1 Result of the chemical and physical analysis of the soil used to fill the experimental units. Piracicaba, SP, Brazil, 2016/17. 

At 60, 45, 30, 15, and 0 days, eight pots were applied at a time, four for each cultivar according to the periods between application and soybean sowing. After the application of the last period, the soybean cultivars BMX Garra RR2/STS and M 6410 IPRO (non-STS) were sown with ten seeds per pot, on January 5, 2017.

The applications were performed with a CO2 pressurized backpack sprayer, bar-equipped with four spray nozzles, at a constant pressure of 200 kPa, a flow rate of 0.65 L min^-1, at the height of 50 cm from the target, and a speed of 1 m s^-1, with an applied band of 50 cm wide by a nozzle, delivering a spray volume of 200 L ha^-1.

At 7, 14, 21, and 28 DAS, soybean seedling emergence was evaluated. At 28 DAS, plant height was measured in all the plants in each pot and then averaging the plant height per pot. Also, at 28 DAS, the aerial part of all soybean plants was collected, the plant material was oven-dried at 65 °C for 72 h, and then the dry mass was recorded. The total dry mass per pot and mean dry mass per plant were determined.

Data were tested by analysis of variance and F-test (P<0.05), according to ^{Pimentel-Gomes and Garcia (2002}). For the soybean cultivar, the means were compared by ^{Tukey's test (P<0.05) (Tukey, 1949}). While for the period, the means were subjected to regression analysis (P<0.05). The Sivar 5.6 program was used for the analysis (^{Ferreira, 2011}).

RESULTS AND DISCUSSION

There is a difference between cultivars in the emergence evaluation at 7 DAS (Figure 1). For the periods 0 and 15 days, between the application and sowing, the cultivar BMX Garra presented superior emergence than the cultivar M 6410 IPRO. For the other periods, no differences were detected between cultivars.

Figure 1 Emergence (%) of soybean plants at 7 DAS, under pre-sowing application of metsulfuron (2.4 g a.i. ha^-1). Piracicaba, SP, Brazil, 2016/17. Means with the same letters, comparing cultivars, do not differ by Tukey test (P<0.05). 

Similarly, differences between cultivars at 0 days were found in the emergence evaluation at 14 DAS (Figure 2). The cultivar STS showed a higher percentage of emergence in this period than the non-STS cultivar; for the other periods, no differences were observed between the cultivars.

Figure 2 Emergence (%) of soybean plants at 14 DAS, under pre-sowing application of metsulfuron (2.4 g a.i. ha^-1). Piracicaba, SP, Brazil, 2016/17. Means with the same letters, comparing cultivars, do not differ by Tukey test (P<0.05). 

For the emergence of STS cultivar, at 7 and 14 DAS, it was possible to fit a linear regression; with the decrease of days in the period between application and sowing, the emergence percentage was lower. For the emergence of non-STS cultivar, at 7 and 14 DAS, it was possible to fit a polynomial. Despite the fit, 80.0% emergence occurs at 7 DAS for the STS cultivar (0 d between application and sowing), which is at the recommended minimum limit for seed commercialization (^{MAPA, 2009}). While for the non-STS cultivar, the emergence percentage at 7 DAS was only 30.0% for the same period.

It should be noted that the emergence evaluation was also performed at 21 and 28 DAS (data not shown). However, they were not subjected to statistical analysis since the percentages remained constant from the evaluation at 14 DAS.

For the height of plants (Figure 3), it can be observed that for all periods, values for the STS cultivar were higher than the values of the non-STS cultivar. For the non-STS cultivar, it was possible to fit an upward linear regression, with the increase of days between the application and sowing, the plants' mean height of the increases. While for the STS cultivar, it was not possible to fit a linear regression according to the observed criteria (biological explanation, significant regression, non-significant deviations from regression, coefficient of determination, and residual analysis).

Figure 3 Height (cm) of soybean plants at 28 DAS, under pre-sowing application of metsulfuron (2.4 g a.i. ha^-1). Piracicaba, SP, Brazil, 2016/17. Means with the same letters, comparing cultivars, do not differ by Tukey test (P<0.05). 

For the variables total dry mass (Figure 4) and mean dry mass (Figure 5), for the STS cultivar, it was not possible to fit a linear regression according to the observed criteria (biological explanation, significant regression, non-significant deviations from regression, coefficient of determination, and residual analysis). It is evidence of the tolerance of the BMX Garra cultivar to the herbicide metsulfuron in pre-sowing, regardless of the period in days, between application and sowing.

Figure 4 Total dry mass (g) of aerial part soybean plants at 28 DAS, under pre-sowing application of metsulfuron (2.4 g a.i. ha^-1). Piracicaba, SP, Brazil, 2016/17. Means with the same letters, comparing cultivars, do not differ by Tukey test (P<0.05). 

Figure 5 Average dry mass (g) of aerial part soybean plants at 28 DAS, under pre-sowing application of metsulfuron (2.4 g a.i. ha^-1). Piracicaba, SP, Brazil, 2016/17. Means with the same letters, comparing cultivars, do not differ by Tukey test (P<0.05). 

However, for the non-STS cultivar, it was possible to fit a linear regression; decreasing the days between application and sowing, the total and mean dry mass of soybean plants also decrease. It should be noted that in the comparison between cultivars, it is not verified difference only for the period of 60 days, for total dry mass. In other periods, the STS cultivar always had the highest total dry mass (Figure 4).

For the mean dry mass of soybean plants, it was observed that, in the comparison between cultivars, the cultivar STS was superior to non-STS for periods of 0, 15, and 30 days between application and sowing. No differences were detected between the cultivars for the periods of 45 and 60 days (Figure 5).

Studies have reported tolerance of STS cultivars for application of the following sulfonylureas: chlorimuron (^{Poston et al., 2008}; ^{Silva et al., 2016}; ^{Albrecht et al., 2017}; ^{Albrecht et al., 2018}) and prosulfuron (^{Anderson and Simmons, 2004}), thifensulfuron (^{Esbenshade et al., 2001}), nicosulfuron (^{Manley et al., 2001}; ^{Silva et al., 2016}; ^{Albrecht et al., 2017}; ^{Silva et al., 2018}; ^{Silva et al., 2019}), halosulfuron (^{Nandula et al., 2009}), trifloxysulfuron ( ^{Porterfield et al., 2006}), sulfometuron (^{Piasecki and Rizzardi, 2016}) and metsulfuron (^{Merotto Júnior et al., 2000}; ^{Albrecht et al., 2017}). Although there are reports of tolerance to herbicide metsulfuron in some STS cultivars, ^{Silva et al. (2016)} did not verify tolerance, for post-emergence application, in the cultivar CD 2630 RR/STS.

For non-STS cultivars, the safety interval between application and sowing is 60 days (^{Rodrigues and Almeida, 2018}). In the present study, the cultivar M 6410 IPRO (non-STS) had its initial development impaired by the pre-emergence application of the herbicide metsulfuron. It was possible to fit a regression model for all analyzed variables, reducing the values with a decrease of the days between application and sowing. According to the results, reductions are observed in some STS's variables from 45 days between application and sowing. For the interval of 60 days, in general, no differences are found between the cultivars, also for 60 days, an emergence of 85 and 95% was observed, at 7 and 14 DAS, respectively, for non-STS. Values above of recommended minimum (80%) for seed commercialization (^{MAPA, 2009}).That is, the pre-sowing application of metsulfuron is safe for the non-STS cultivar when the safety interval is respected.

In pre-emergence, there are no reports of selectivity of the herbicide metsulfuron in STS soybean, the results obtained in the present study are of great importance in the positioning of the herbicide metsulfuron in pre-emergence of STS soybean. The results indicate that the cultivar BMX Garra RR2/STS is potentially tolerant of metsulfuron, regardless of the period between application and sowing. Only for the emergence, it was possible to fit a regression; however, even for sowing immediately after application, the emergence percentages were 80.0%.

The herbicide metsulfuron has a broad spectrum of action, controlling mainly eudicotyledons plants and some species of monocotyledons (^{Rodrigues and Almeida, 2018}). The herbicide, alone or in combination, is effective in the control of Conyza bonariensis (^{Vargas et al., 2007}; ^{Walker et al., 2013}), Fimbristylis miliacea (^{Tormena et al., 2016}), Hordeum spontaneum (^{Izadi-Darbandi and Aliverdi, 2015}), Cirsium arvense (^{Zargar et al., 2019}), among others.

Thus, the herbicide metsulfuron can be used in the management of weeds, especially eudicotyledons plants resistant or naturally tolerant to glyphosate in pre-sowing of STS soybean. In this context, STS cultivars can be of great importance in the management and prevention of the selection of herbicide-resistant biotypes weeds, due to the possibility of using other herbicides (Green, ²⁰⁰⁷; ²⁰¹²). As well as the use of the herbicide metsulfuron, in pre-emergence, according to these results. Studies indicate that the use of pre-emergence herbicides, in combination with other management practices, is effective in the control of weeds in soybean crops, as well as in the prevention of the selection of resistant weed biotypes (^{Neve et al., 2018}; ^{Rosario-Lebron et al., 2019}).

CONCLUSIONS

The soybean cultivar BMX Garra RR2/STS, in general, was potentially tolerant to the herbicide metsulfuron, applied at pre-sowing. Therefore, sowing can be done immediately after the application. The cultivar M 6410 IPRO (non-STS) had its initial development impaired by the pre-emergence application of the herbicide metsulfuron. However, when the safety interval recommended (60 days) is respected, the application of the herbicide is safe to the cultivar.