Onion (Allium cepa) is the third most economically important vegetable produced in Brazil (IBGE 2016). Downy mildew of onion, caused by Peronospora destructor, has widespread occurrence and can affect onion leaf area severely, leading to yield reduction (Scholten et al. 2007). The disease can cause yield losses of up to 75%, under favorable conditions (Develash and Sugha 1997). Onion production is concentrated in the southern region of Brazil, with 45.6% of production and approximately 678,000 tons grown annually (IBGE 2016). To control downy mildew in this region, systemic and protective fungicides are often applied, as frequently as every seven or fourteen days.

Plant activators have been used as a control alternative or integrated with fungicides to manage diseases of vegetables. Acibenzolar-S-methyl (ASM), an inducer of systemic acquired-resistance in plants, has been effective in inducing resistance in Datura stramonium and Nicotiana benthamiana against Iris yellow spot virus (Tripathi and Pappu 2015), as well as in onion against Xanthomonas axonopodis pv. allii (Lang et al. 2007). Potassium phosphite was found to provide protection against potato late blight (Phytophthora infestans) in field trials (Liljeroth et al. 2016). However, frequent applications of ASM may negatively affect the yield of tomatoes (Kunwar et al. 2017). Therefore, the aim of the present work was to evaluate the efficiency of calcium phosphite, potassium phosphite, and acibenzolar-S-methyl, using different doses and application intervals, on the management of downy mildew of onion under field conditions.

The field trial was conducted in 2015 at Epagri/experimental station of Ituporanga, Santa Catarina state, in the south region of Brazil. The cultivar used was Empasc 352 – Bola Precoce, which is susceptible to downy mildew (Araújo et al. 2017). Onion seedlings were grown on soil beds during April. The seedlings were transplanted to the field approximately 70 days after sowing (during July). All cultural practices were performed as the standard onion production system recommended by Santa Catarina, with soil pH ~ 6; 100 kg of N/ha, 80 kg of P/ha, and 60 Kg of K/ha. The plants were not inoculated and the presence of pathogen structures such as sporangia and sporangiophores were observed under a stereoscopic loupe after 60 and 80 days of transplanting to confirm natural infection. Sporangia and sporangiophores will be incorporated into the work collection of phytopathogenic fungi of Phytosanitary Laboratory at Epagri/EEITU (code EITF 20), according to the preservation technique of freeze-drying (Laviola et al. 2006).

The products and doses used are described in Table 1. Each chemical was sprayed in the field at weekly intervals (seven, 14 and 21 days) starting 28 days after transplanting. Water treated plants were used as a control. The treatments were applied manually using pressure sprayers (SeeSa®, 2-l capacity), corresponding to spray volume of 500 L per hectare.

Table 1 Area under disease progress curve (AUDPC) and commercial yield in plots treated with different products, doses and application intervals, in Santa Catarina state, Brazil, to control downy mildew of onion (Peronospora destructor)
Full size table

The experimental plots were composed of 120 plants (six rows with 20 plants), arranged in a spacing of 0.35 m (between rows) × 0.10 m (between plants), simulating a density of approximately 300,000 plants per hectare. Disease severity was assessed weekly for six weeks starting 42 days after transplanting. The severity was rated for each plot using a rating description scale (1–9), which assigns notes to the entire plot (Mohibullah 1991). The values from scale notes of the experimental plots were used to calculate the area under disease progress curve – AUDPC (Shaner and Finney 1977). We also evaluated the incidence of thrips (Thrips tabaci Lindeman), one of the major pests of onion in Brazil (Gonçalves 2006), in the central leaves of five plants which were randomly chosen per plot at 81 and 88 days after transplanting. For this, the nymphs population was assessed visually (0 = absence of nymphs; 1 = up to 6 nymphs; 3 = up to 15 nymphs; and 9 ≥ 20 nymphs). Onion bulbs were harvested at 122 days after transplanting. Bulbs were harvested from the central rows of each plot, discarding single border rows. For yield calculation, we considered only the commercial bulbs (bulbs with less than 35 mm in diameter, and rotten bulbs were not considered).

The experiment was arranged in a completely randomized block design using three replications per treatment, in the factorial design 3 × 3 × 3 (product × dose × application interval), and one additional treatment (non-treated control). The values of AUDPC and commercial yield were analysed using ANOVA and, whenever necessary, means were grouped with Scott-Knott cluster analysis (p = 0.05). The GENES® (Cruz 2016) software was used for all analyses.

No significant effects of the treatments or their interactions were observed for the yield: product (p-value = 0.36); dose (p-value = 0.42); application interval (p-value = 0.42); product × dose (p-value = 0.32); product × application interval (p-value = 0.65); dose × application interval (p-value = 0.40); product × dose × application interval (p-value = 0.82). For AUDPC, the only significant effect was observed for the interaction product × dose (p-value = 0.03), but its combinations did not present significant difference compared to the control. The other treatments and their interactions showed no significant effect: product (p-value = 0.22); dose (p-value = 0.25); application interval (p-value = 0.25); product × application interval (p-value = 0.79); dose × application interval (p-value = 0.39); product × dose × application interval (p-value = 0.35). However, the overall mean of AUDPC (145.45, with p-value = 0.44) and yield (18.40, with p-value = 0.76) from treatments did not differ from the mean of the non-treated control (Table 1). Thrips incidence scores were statistically similar for all treatments according to the Friedman test.

The efficiency of acibenzolar-S-methyl and phosphites on the management of onion diseases has presented variable results. Wordell Filho et al. (2007) evaluated the effect of weekly applications of acibenzolar-S-methyl (10 g of active ingredient per hectare) and potassium phosphite (Phyto’s K®, 1 l of commercial product per hectare) on the control of downy mildew of onion in Santa Catarina, Brazil, and they also observed that disease severity and yield were similar to the non-treated control. On the other hand, for the control of downy mildew of onion in Colombia, applications (eight-day intervals) of potassium phosphite alone or alternately with metalaxyl/mancozeb resulted in a yield up to 55% higher than the control (Monsalve et al. 2012). For the xanthomonad leaf blight of onion (Xanthomonas axonopodis pv. allii), four sprays of acibenzolar-S-methyl had the same level of control as nine sprays of copper hydroxide/mancozeb, however, without increasing yield (Gent and Schwartz 2005). Difference in temperature, relative humidity, plant density, level of waxy leaf, pathogen types, and local inoculum pressure are some factors that may explain these discrepancies in the results from different regions.

According to the present study, acibenzolar-S-methyl, and calcium and potassium phosphites at rates and frequency used in this study are not effective in controlling downy mildew of onion. Therefore, tests with other inducers (Beta-aminobutyric acid) will be started with the purpose of rationing the use of fungicides.