Response of Four Dry Bean Market Classes to Pre-Emergence Applications of Pyroxasulfone, Sulfentrazone and Pyroxasulfone plus Sulfentrazone

Only one herbicide mode of action (ALS inhibitor) is currently available to Ontario dry bean producers for soil-applied broadleaf weed control. Four field studies were conducted over two years (2014, 2015) to examine the tolerance of four market classes of dry beans to sulfentrazone (210 and 420 g∙ai∙ha−1) and pyroxasulfone (100 and 200 g∙ai∙ha−1) applied alone and in combination. The registration of these two herbicides would provide Ontario dry bean producers with two additional modes of action for broadleaf weed control. Pyroxasulfone caused up to 23%, 6%, 7% and 10% injury in adzuki, kidney, small red Mexican and white bean, respectively; sulfentrazone caused up to 51%, 12%, 15% and 44% injury and the combination caused up to 90%, 23%, 29% and 62% injury, respectively. Kidney and small red Mexican bean density, height, seed moisture content and yield were not affected. Pyroxasulfone (200 g∙ai∙ha−1) + sulfentrazone (420 g∙ai∙ha−1) reduced adzuki and white bean density, shoot dry weight, height and yield. This study concludes that pyroxasulfone (100 g∙ai∙ha−1) + sulfentrazone (210 g∙ai∙ha−1) applied PRE can be safely used to control weeds in Ontario kidney and small red Mexican bean production.


Introduction
In 2015, Ontario dry bean growers produced 115,000 tonnes of white and coloured dry beans (Phaseolus vulgaris L.) on 52,600 ha [1].Dry bean market classes commonly grown in Ontario include black, cranberry, kidney, otebo, small red Mexican, white (navy), and adzuki (Vigna angularis (Willd) Ohwi & Ohashi) bean.One of the most critical aspects of dry bean production is weed control, as weed interference has been reported to cause up to 68% yield loss [2]- [4].Currently, Ontario dry bean producers are limited to two acetolactate synthase (ALS) inhibitor herbicides for soil-applied broadleaf weed control, to which there are currently more than ten resistant weed species in the province of Ontario [5].Therefore, another mode of action is needed for soil-applied broadleaf weed control in order to chemically control these weeds.
Sulfentrazone is a protoporphyrinogen oxidase (PPO) inhibitor herbicide that is recently registered for use in Ontario soybean (Glycine max L.).Sulfentrazone inhibits the PPO enzyme, causing an excess of protoporphyrinogen IX in the chloroplast of the cell [6].The excess then leaks out of the chloroplast into the cyto-phyrinogen, where various reactions convert it to free oxygen radicals.These radicals disrupt cell membranes, resulting in cellular leakage, loss of cell function, and cell death [7].Sulfentrazone controls broadleaf weeds such as pigweeds (Amaranthus spp.), common lambsquarters (Chenopodium album L.) and wild buckwheat (Polygonum convolvulus L.), and grasses such as giant foxtail (Setariafaberii Hermm.), fall panicum (Panicum dichotomiflorum Michx.) and witchgrass (Panicum capillare L.) [8]- [11].
There is little information on dry bean tolerance to sulfentrazone, pyroxasulfone or their combination.The objective of this study was to determine the tolerance of four market classes of dry beans to PRE applications of sulfentrazone, pyroxasulfone, and a combination of the two.

Materials and Methods
Four field trials were conducted at the University of Guelph Ridgetown Campus, Ridgetown and Huron Research Station, Exeter in Ontario, Canada over two years (2014,2015).The 2014 Ridgetown site was a loamy soil with sand, silt and clay content of 40%, 35% and 25%, respectively, pH of 6.6 and organic matter content of 7.1%.The 2014 Exeter site was a Brookston clay loam, with 41% sand, 40% silt, 19% clay, pH of 7.8 and organic matter content of 3.8%.In 2015, soil at Ridgetown was a sandy clay loam with 53% sand, 27% silt, 20% clay, pH of 7.0 and organic matter content of 5.3%.The soil at Exeter in 2015 was a silty clay loam with 18% sand, 46% silt, 36% clay, pH of 7.5 and organic matter content of 4.7%.Seedbeds were prepared by an autumn moldboard ploughing followed by two passes in the spring with a cultivator with rolling basket harrows.
Experiments were arranged in split-plot design, with herbicide as the main plot and bean market class as the sub-plot.Bean market classes included adzuki ("Erimo"), kidney ("Red Hawk"), small red Mexican ("Merlot"), and white ("T9905") bean.Each experiment had four replicates.Plots in Ridgetown were 6 m wide by 8 m long, and consisted of two rows of each market class spaced 75 cm apart.Plots in Exeter were 3 m wide by 10 m long and consisted of 1 row of each market class spaced 75 cm apart.In Ridgetown, kidney bean was seeded at approximately 175,000 seeds ha −1 while adzuki, white and small red Mexican beans were seeded at approximately 235,000 seeds ha −1 .In Exeter, kidney and small red Mexican beans were seeded at 200,000 seeds ha −1 , adzuki bean was seeded at 275,000 ha −1 , and white bean was seeded at 250,000 seeds ha −1 .All experiments were seeded 4 to 5 cm deep between late May and early June.Herbicide treatments (based on manufacturer's recommended rate) included an untreated control, pyroxasulfone at 100 and 200 g•ai•ha −1 , sulfentrazone at 210 and 420 g•ai•ha −1 , pyroxasulfone (100 g•ai•ha −1 ) + sulfentrazone (210 g•ai•ha −1 ), and pyroxasulfone (200 g•ai•ha −1 ) + sulfentrazone (420 g•ai•ha −1 ).Herbicides were applied with a CO 2 pressurized backpack sprayer calibrated to deliver 200 L ha −1 at 240 kPA using a 1 m handheld boom with three ULD 110-02 nozzles (Hypro, New Brighton, MN) spaced 0.5 m apart.Plots were not irrigated but were fertilized according to recommended Ontario agronomic crop production practices.All plots were maintained weed-free the entire season by hand-hoeing to eliminate the confounding factor of weed interference.
Visible crop injury was assessed at 1, 2, 3, 4, 6 and 8 weeks after crop emergence (WAE).Plant density and shoot dry weight were determined at 3 WAE by counting the number of plants in a metre of row for each market class in each treatment, clipping them at the soil surface, placing them in a separate paper bag, drying in a kiln and weighing once dry.At 6 WAE, plant height was determined by randomly selecting ten plants in each market class in each treatment and averaging their heights.Seed moisture content was determined at maturity.Yield was determined at maturity and adjusted to 15% moisture for adzuki bean and 18% moisture for kidney, small red Mexican and white bean.

Statistical Analysis
Statistical Analysis Software (SAS) v. 9.4 (SAS Institute Inc., NC) was used to perform an analysis of variance on all data.Data were partitioned into the fixed effects of herbicide treatment, bean market class and their interaction, and the random effects of environment (year-location combination), block within environment, and interactions between environment and each of the fixed effects.The significance of the fixed and random effects was determined using the F-test and Z-test, respectively.The UNIVARIATE procedure was used to produce residual plots and Shapiro-Wilk statistics to analyze the homogeneity and normality of the residuals under various transformations.The transformation that produced the highest Shapiro-Wilk statistic was used to analyze each data set.Injury, biomass, and seed moisture data were log transformed, while yield and height data were arcsine square root transformed.The MIXED procedure was used to obtain the least squares means and differences of least squares means to achieve means separation between treatments at a Type 1 error of 0.05.Orthogonal contrasts were also performed.Data were back transformed for presentation on the original scale.

Results and Discussion
Statistical analysis determined there was no interaction between bean market class and herbicide treatment for injury at 1 and 2 WAE (Table 1); therefore the main effect of herbicide will be discussed separately.There was an interaction between dry bean market class and herbicide treatment for plant density, shoot dry weight, plant height, seed moisture content and yield (Table 2).Therefore, only the interaction effects will be discussed.
At 3, 4, 6 and 8 WAE, injury was greatest in adzuki bean, followed by white, small red Mexican, then kidney bean (Table 1).Pyroxasulfone (100 g•ai•ha −1 ) produced the least amount of injury of the herbicide treatments, causing 5% injury or less in all market classes (Table 3).Pyroxasulfone (200 g•ai•ha −1 ) caused greater injury than the 100 g•ai•ha −1 rate in adzuki bean at 4 and 6 WAE.Pyroxasulfone (200 g•ai•ha −1 ) caused up to 23%, 6%, 7%, and 10% injury in adzuki, kidney, small red Mexican and white bean, respectively.The level of injury observed in adzuki and white bean is consistent with other studies examining pyroxasulfone in dry bean, but the injury observed in kidney and small red Mexican bean were much lower than the 10% to 21% injury caused by pyroxasulfone (209 g•ai•ha −1 ) in other studies using the same cultivars [17] [18].Weather conditions (rainfall, temperature, etc.) and soil factors affecting herbicide availability may have contributed to the differences among studies.
Sulfentrazone (210 g•ai•ha −1 ) caused up to 17% injury in adzuki bean, 4% injury in kidney and small red Mexican bean, and 16% injury in white bean (Table 3), which is consistent with a study by Soltani et al. [19].The high rate of sulfentrazone (420 g•ai•ha −1 ) caused greater injury than the 210 g•ai•ha −1 rate in small red   Mexican bean at 4 WAE, but injury was equivalent for both rates in the other market classes.Sulfentrazone (420 g•ai•ha −1 ) caused up to 51%, 12%, 15% and 44% injury in adzuki, kidney, small red Mexican and white bean, respectively.Pyroxasulfone (100 g•ai•ha −1 ) + sulfentrazone (210 g•ai•ha −1 ) caused up to 41, 6, 5, and 12% injury in adzuki, kidney, small red Mexican and white bean, respectively, and pyroxasulfone (200 g•ai•ha −1 ) + sulfentrazone (420 g•ai•ha −1 ) caused up to 90, 23, 29, and 62% injury, respectively (Table 3).In other studies examining PRE tank mixtures of a group 14 plus a group 15 herbicide on dry bean, dimethenamid-p/saflufenacil at the 1X rate caused up to 18% injury in dry bean, while the 2X rate caused up to 45% injury [20].Fomesafen + s-metolachlor only caused up to 6.4% injury [21].With the exception of white bean at 6 WAE, the level of injury produced by either herbicide combination was not greater than the injury produced by the respective rate of sulfentrazone on its own.Based on orthogonal contrasts, the addition of sulfentrazone to pyroxasulfone caused an increase in bean injury of 5% to 18% at 1, 2, 3, 4, 6 and 8 WAE (Table 1).In contrast, the addition of pyroxasulfone to sulfentrazone caused an increase in bean injury of 5% to 12% at 1, 4 and 6 WAE.

Plant Density
Kidney bean and small red Mexican bean densities were not reduced by any of the herbicide treatments (Table 4).Pyroxasulfone (100 and 200 g•ai•ha −1 ), sulfentrazone (210 and 420 g•ai•ha −1 ), and pyroxasulfone (100 g•ai•ha −1 ) + sulfentrazone (210 g•ai•ha −1 ) did not reduce adzuki or white bean density relative to the control.In Table 4. Plant density, shoot dry weight and plant height (expressed as percentages of the untreated control) of four market classes of dry bean treated PRE with sulfentrazone and pyroxasulfone, alone and in combination, near Ridgetown and Exeter, ON, Canada over two years (2014,2015).Means followed by the same letter in a row (W-Z) or column (a-c) for each variable are not significantly different according to Fisher's Protected LSD at P < 0.05.contrast, pyroxasulfone (200 g•ai•ha −1 ) + sulfentrazone (420 g•ai•ha −1 ) reduced adzuki and white bean density by 45% and 33%, respectively.Dimethenamid-p/saflufenacil also reduced plant density by 49% [20].Based on orthogonal contrasts, the addition of sulfentrazone to pyroxasulfone reduced dry bean density by 2 plants m −1 of row; however, there was no impact on dry bean density when pyroxasulfone was added to sulfentrazone (Table 2).

Seed Moisture Content
Elevated seed moisture content indicates a delay in crop maturity, which will cause increased drying costs.In this study, kidney, small red Mexican and white bean seed moisture content was not affected by any of the herbicide treatments compared to the control (Table 5).Pyroxasulfone (200 g•ai•ha −1 ) + sulfentrazone (420 g•ai•ha −1 ) increased adzuki bean seed moisture by 2.4%.Given that the seed moisture content of adzuki bean treated with sulfentrazone (420 g•ai•ha −1 ) was not different from this combination, the delayed maturity can be attributed to the sulfentrazone rather than the pyroxasulfone.This is reflected by the orthogonal contrasts, which show the addition of sulfentrazone to pyroxasulfone increased bean moisture content (delayed maturity) at harvest by 0.9%; however, there was no impact on bean maturity when pyroxasulfone was added to sulfentrazone (Table 2).

Conclusion
In general, the level of injury caused by pyroxasulfone was lower and sulfentrazone was higher in this study than that reported in other studies, which might be a result of differential weather and soil conditions.Adzuki bean was the least tolerant to PRE mixtures of pyroxasulfone + sulfentrazone, with up to 90% injury and a 59% yield loss. White bean was the second most sensitive, with a narrow margin of crop safety at the low rates, and kidney and small red Mexican bean had acceptable tolerance at the low rates.Based on the results of this study, pyroxasulfone (100 g•ai•ha −1 ) + sulfentrazone (210 g•ai•ha −1 ) applied PRE is a safe herbicide combination for kidney and small red Mexican beans in Ontario.

Table 1 .
Visible injury of four market classes of dry bean due to pyroxasulfone and sulfentrazone, alone and in combination, applied PRE in 2014 and 2015 at the Huron Research Station, Exeter, ON and University of Guelph Ridgetown Campus, Ridgetown, ON.
a Abbreviations: B, bean market class; H, herbicide treatment; NS, not significant at α = 0.05; PRE, pre-emergence; WAE, weeks after emergence.b Means of main effects were only separated when the interaction was not significant.Significance of main effects, interactions and contrasts are denoted by * for P < 0.05 and ** for P < 0.01.b Means followed by the same letter in each column are not significantly different according to Fisher's Protected LSD at P < 0.05.

Table 2 .
Plant density, shoot dry weight, plant height, seed moisture content and yield of four market classes of dry bean treated PRE with sulfentrazone and pyroxasulfone, alone and in combination, near Ridgetown and Exeter, ON, Canada over two years(2014, 2015).Significance of main effects, interactions and contrasts indicated by * for P < 0.05 and ** for P < 0.01.Means were not separated as the interaction was significant. a

Table 3 .
Visible injury (%) of four market classes of dry bean treated PRE with sulfentrazone and pyroxasulfone, alone and in combination, near Ridgetown and Exeter, ON, Canada over two years(2014, 2015).Means followed by the same letter in a row (Y-Z) or column (a-d) for each rating are not significantly different according to Fisher's Protected LSD at P < 0.05 a .

Table 5 .
Seed moisture content (%) and yield (as a percent of the untreated control) of four market classes of dry bean treated PRE with sulfentrazone and pyroxasulfone, alone and in combination, near Ridgetown and Exeter, ON, Canada over two years(2014, 2015).Means followed by the same letter in a row (Y-Z) or column (a-c) are not significantly different according to Fisher's Protected LSD at P < 0.05.