The arid and semiarid areas of Mexico account for over 50% of the country’s area and they are divided into scrublands (85%) and grasslands (15%), as opposed to 150 years ago (^{PMARP, 2012}). Due to this, ^{Sánchez et al. (2018}) point out the importance of the native grasses to recover the older condition and functionality of grass without altering the vulnerable arid ecosystem. To establish rainfed prairies using seeds (sensu lato), there are recommendations such as using complete propagules (glumes, grasses, palea, awns and modified twigs) or caryopses, which will have a larger embryo, more endosperm, seedling weight and vigor, the larger it is (^{Quero et al., 2016}; ^{Quero et al., 2017}). However, one possible problem one may come across is the pathogenic fungi related to the seeds of important forage species that may result in cattle diseases (^{Pirelli et al., 2016}), therefore one infected and/or contaminated seed may introduce a pathogen in a plot/region/country (^{Sandoval et al., 2012}). Seed treatments are an efficient tool against the negative impacts of diseases, and they help farmers produce higher-quality crops (^{FAO and AfricaSeeds, 2019}). Eradicating treatments are more specialized than preventive ones, and they are designed to eliminate a specific pathogen by physical or chemical means, and can be effective against states of profound infection, since they can penetrate seed tissue and kill pathogens without causing phytotoxicity. The biological control of seed-transmitted pathogens is performed via three mechanisms: resistance induction, competition or elimination of the pathogen and the production of antibiotics, and they are based on the antagonism that microorganisms may have with each other (^{Maude, 1985}). In the market there are contact agrochemicals, which persist on the plant’s exterior (foliar application) and stop the fungi from germinating and penetrating the crop’s cells, a few of which include include Captan, Zineb, Maneb, Mancozeb, Thiram, Folpet, Quintozane and Clorotalonil (^{Chirinos et al., 2020}). Another group, the systemic agrochemicals, are absorbed via the foliage, stems and roots and where the vascular system helps spread the active compounds throughout the plant (^{Arriagada, 2000}). The use of agrochemicals entails the knowledge of its mechanism of action to avoid the risk of resistance, therefore knowledge on the sensitivity of the pathogen to the fungicide is vital for the adequate sanitary control of the seed (^{Sandoval et al., 2012}). The most widely used method to determine sensitivity is to integrate a pesticide product, under controlled levels and laboratory conditions, into the culture medium (^{Dhingra and Sinclair, 1995}). In Mexico, the need for forage genetic resources that guarantee the establishment, quality, high persistence and adaptation to regional conditions makes studies on seed health crucial (^{Quero et al., 2007}), as well as to know the sensitivity of agrochemicals in order to integrate the management of seed health. Therefore, the aim of this investigation was to determine the in vitro effect of six fungicides and one biological one (Bacillus subtilis) against Alternaria alternata , Bipolaris cynodontis and Fusarium incarnatum.

A. alternata, B. cynodontis and F. incarnatum cultures were used, isolated from commercial B. curtipendula seeds, harvested in 2017 (^{Quero et al., 2020}). The cultures were planted in a PDA (potato-dextrose-agar) medium for eight days, at 28 ± 2 °C. Commercial fungicides were evaluated: contact (Captan and Mancozeb), systemic (Benomyl, Prochloraz, Thiabendazole and Thiophanate-methyl) and a B. subtilis based biological product. Final concentrations were calculated based on the active ingredient and the volume of PDA medium to be prepared (Table 1). The effect of the product was determined with the aggregation of the fungicide to the PDA culture medium (^{Dhingra and Sinclair, 1995}) at different concentrations: 0 (control), 0.005, 0.001, 0.05, 0.01, 0.5, 0.1, 1, 5, 10, 100, 150, 200, 250 and 300 mg L^-1. A mycelium disk, 0.5 cm in diameter, was planted in the center of Petri dishes, with three repetitions per concentration. The disks were obtained from pure cultures with a growth of six days in PDA and incubated in the dark at 25 ± 2 °C. The bidirectional diameter of cultures was measured every 48 h, until the control filled the dish, and with the average, the mycelial growth was calculated:

%IM=CML-CMICMLx 100

where: IM, inhibition of mycelial growth (%); CML, Mycelial growth; CMI, Influenced mycelial growth (^{Patiño and Rodríguez, 2001}).

A completely random design was used and, using the data on inhibition percentages, a simple analysis of variance was carried out for each pathogen. Since the normal distribution of data was not carried out, these were transformed by the arcsine (^{Gabriel et al., 2021}) and the comparison of averages using Tukey’s test (p≤0.05) using the statistics program R. The effective concentration (CE₅₀) was obtained by transforming each concentration [log(x)] and relating it with the percentage of inhibition.

In most treatments, the mycelial growth of the pathogens was reduced in comparison with the control. Both the sensitivity of A. alternata, B. cynodontis and F. incarnatum to the different concentrations of the fungicides evaluated, and the calculation of the highest effective concentration that inhibits 50% of the mycelial growth of the fungus (CE₅₀) (Table 2), were determined eight days after planting.

Effect on the mycelial growth of Alternaria alternata . Thiabendazole and Benomyl in different concentrations were not efficient in inhibiting the fungus, since a maximum inhibition of 7.2 and 10.8% were observed with higher doses and 5 and 100 mg L^-1, respectively (Table 2). These data coincide with ^{Herrera et al. (2011}) and ^{Cristóbal et al. (2013}), who indicated that, with doses of 500 and 450 mg L^-1 of Benomyl, 35.6 and 45% of the growth of Alternaria spp. is inhibited; these concentrations are higher than those evaluated in this experiment. In the case of Captan, an inhibition of 83.3% was observed at 200 mg L^-1, higher than what was reported by ^{Parveen et al. (2013}) with an inhibition of 36% of A. alternata at the same concentration. Mancozeb displayed similar data to Captan with an 85% inhibition, but at a higher dose (Table 2). ^{Malandrakis et al. (2015}) indicated similar data for A. alternata. Prochloraz displayed a similar inhibition rate to Mancozeb and Captan, with 84.1%, but at a lower dose (1 mg L^-1), and therefore this product may be feasible for the control of this fungus. ^{Iacomi et al. (2004}) report an inhibition of 100% for A. alternata in radish plants, higher than the data reported for this work. Only Thiophanate-methyl (10 mg L^-1) produced an inhibition of 100% at the highest dose. B. subtilis, starting at 0.05 and 1 mg L^-1, produced an inhibition of 93%, similar to reports by ^{Ñacato et al. (2018}), who determined B. subtilis to be highly efficient for the biological control against Alternaria spp. (Table 2).

Effect on the mycelial growth of Bipolaris cynodontis. Thiabendazole (5 mg L^-1) and Benomyl (100 mg L^-1) displayed the lowest efficiency in the inhibition of the mycelial growth of B. cynodontis, with inhibitions of 4.5 and 27.7%. ^{Alburqueque and Gusqui (2018}) reported an inhibition of 10.3% on Phytophthora infestans; 37%, for Botrytis cinerea and 100% for Rhizoctonia solani, at 200 mL L^-1 of Thiabendazole, with a higher dose in comparison with this investigation.

Captan, Mancozeb, Prochloraz and Thiophanate-methyl inhibited the fungus by 100% at high concentrations, which was lower than reports by ^{Imran et al. (2013}), who registered an inhibition of 53% of B. oryzae at 50 mg L^-1. ^{Arce et al. (2019}) pointed out that the percentages of inhibition of Bipolaris spp. with Benomyl, Mancozeb and Thiophanate-methyl were low in comparison with the values observed in this investigation (Table 2). ^{Rondón et al. (2006}) report similar data for inhibition (100%) in Colletotrichum gloeosporioides with Prochloraz and starting at 100 mg L^-1.

For the particular case of the biological product B. subtilis, this was more efficient than the chemical fungicides at the concentrations evaluated, with a percentage of mycelial inhibition between 94.7 and 100%. The data are similar to those reported by ^{Rivero et al. (2008}). For B. subtilis, a CE₅₀ of 0.00023 mg L^-1 is reported. This concentration is lower than the rest of the fungicides evaluated (Table 3).

Effect on the mycelial growth of Fusarium incarnatum. Thiophanate-methyl displayed a lower efficiency of mycelial inhibition for the fungus in the highest dose, with 88.5%. The rest of the fungicides displayed an efficiency of mycelial inhibition between 93 and 100% in the highest doses evaluated. These data show the susceptibility of the fungus to the fungicides evaluated. Regarding CE₅₀, Mancozeb requires high doses (300 mg L^-1) to inhibit 50% of the fungus, therefore its efficiency was low. On the other hand, B. subtilis (CE₅₀ at 0.00014 mg L^-1) was efficient in all doses evaluated, with inhibitions ranging from 97 to 100%, showing that, starting at 0.01 mg L^-1, it inhibited F. incarnatum by 100%. ^{Romero et al. (2015}) obtained similar data with B. subtilis at doses of 0.01 mg L^-1 against F. solani isolated from chayote. On the other hand, ^{Song et al. (2014}) observed that at a concentration of 1 x 10⁶ UFC of B. subtilis, a control of 70% was reached against F. incarnatum, isolated from ginseng (Panax ginseng) roots.

Captan inhibited an average of 93% of the mycelial growth of the three fungi (Table 3), due to its multisite action that interferes with the cell respiration mechanism, making mycelial development difficult and translocating to different tissues to seeds or soil from the treatment (^{Peláez et al., 2016}). Mancozeb presented an efficiency of 93%. Its mechanism of action is based on the modification and inactivation of proteins sensitive to redox, such as those for transcription, translation, and DNA oxidative stress, as well as other metabolical processes that result in cytotoxicity and with a multisite action that generates no resistance in the pathogen (^{Roede and Miller, 2014}). Benomyl was only effective for F. incarnatum, which, like Thiabendazole, was effective only for Fusarium; this was also observed with Benomyl, since they belong to the same chemical group and have similar mechanisms of action (^{Peláez et al., 2016}). ^{Sandoval et al. (2011}) mentioned that Fusarium is sensitive to Benomyl. However, it is likely to be mutagenic and may increase the resistance of pathogens to its effect. On the other hand, Prochloraz displayed an effectiveness of 94%. This fungicide belongs to the chemical group of the Imidazoles, with a preventive, interlaminar, systemic, and curative action that inhibits the biosynthesis of ergosterol in the fungal cell membrane, which also relates to cell growth and division (^{Tapia, 2005}). Thiophanate-methyl displayed a high effectiveness in the inhibition (96%) of growth in the three species of fungi investigated. This is a systemic fungicide (with movement in the plant through the xylem), of the preventive type, which causes abnormalities in the germination of spores, interferes in mitosis and the DNA synthesis of fungal cells (^{Alburqueque and Gusqui, 2018}). The treatment on seeds is known to control root pathogens, such as Rhizoctonia and Fusarium spp., although the best inhibitor was the biological inhibitor B. subtilis, since the minimum concentration evaluated (0.005 mg L^-1) inhibited 97% of the mycelial growth of the three fungal species of phytopathogenic fungi isolated earlier and defined as dominant in commercial B. curtipendula seeds. B. subtilis has been identified as a producer of a wide range of bioactive compounds that potentially inhibit the growth of phytopathogenic fungi (^{Bottero et al., 2017}), which have several action mechanisms, altering cell processes such as the intracellular calcium homeostasis, energetic metabolism, and the processing of RNA (^{Villareal et al., 2018}).

The evaluated concentrations are represented by their logarithmic analog, resulting in the sigmoidal line in which B. subtilis establishes a clear antagonistic response and has a high resistance to handling and easy acquisition in the market, making it a valuable tool to optimize the establishment of prairies of this species, thus improving the seed’s health safety.

The results show the effectiveness of the fungicides evaluated for the control of the three fungi, with the exception of Benomyl and Thiabendazole. Combining contact + systemic fungicides is a common practice that supplements the treatment, ensuring the protection of the seed and its germination by suppressing the action of phytopathogenic fungi (^{Arriagada, 2000}).

The contact fungicides Captan and Mancozeb and systemic fungicides Prochloraz and Thiophanate-methyl gave variable results for the control of A. alternata, B. cynodontis and F. incarnatum. The systemic fungicides inhibited the mycelial growth at low concentrations in comparison with those reached by Captan and Mancozeb. The use of Benomyl and Thiabendazole is not recommendable due to the low inhibition displayed. Bacillus subtilis displayed the highest mycelial growth, starting at 90% for the lowest concentration evaluated (0.005 mg L^-1), therefore representing an important alternative for the improvement of the sanitary quality of the Bouteloua curtipendula seogens, due to its low cost and availability in the market.