The association between fungi and seeds can severely affect seed quality, reducing germination, vigor, seedling emergence and productive potential. It is not always possible to obtain seed lots 100% guaranteed free of pathogens. It is also not possible to ensure that the sown soil or substrate will be clear of fungi. Therefore, seed treatment is advisable in most cases, especially for vegetable hybrid seeds; because treatment cost is very low compared to the high price this type of seeds carries (Cardoso et al., 2015). Seed treatment has been efficient in preventing plant disease outbreaks caused by pathogenic agents in seeds, particularly fungal agents. Chemical seed treatment aims to eradicate these pathogens and/or to protect against soil pathogens, especially during germination. Furthermore, seed treatment can help reduce the volume of fungicides needed to control the diseases. Seed treatment can eliminate the need of foliar use of fungicide products in field crops (Mancini and Romanazzi, 2014). It must be emphasized that beyond the control of pathogens and/or protection of the seeds, the treatment should not impair seed physiological quality (Cardoso et al., 2015). Seed treatment effectiveness depends on, among other factors, seed species and vigor, which may vary from lot to lot (Menten and Moraes, 2010). According to Menten and Moraes (2010), there were 19 fungicides active ingredients registered for treating seeds in Brazil, although this registration is specific for only some species. One of these active substances is Thiabendazole, a member of the Benzimidazole chemical group, registered in Brazil for seed treatment in only few cultures. According to the Ministry of Agriculture, Livestock and Supply (MAPA, 2014), the commercial product Tecto® SC, active ingredient Thiabendazole, is a systemic fungicide registered for soybean and sunflower seed treatment; for avocado, banana, citrus, papaya, mango and melon fruits postharvest treatment and aerial part of avocado, pineapple, coconut, pea, snap pea, papaya, mango, passion fruit, melon and sweet pepper crop treatment. Thus, considering the Cucurbitaceae family, this fungicide has only been registered for treating melon culture, to the control of fungi Colletotrichum orbiculare and Didymella bryoniae, and recommended for spraying plants or postharvest treatment in fruits. Therefore, this product has not been registered for vegetable seed treatment. It has been reported that this fungicide does not impair seed physiological quality of several species, such as alfalfa (Mendes et al., 2001), lentil (Chang et al., 2008), maize (Fessel et al., 2003; Carvalho et al., 2004), soybean (Pereira et al., 2011) and wheat crops (Garcia Júnior et al., 2008). In regard to melon, no studies on the use of this fungicide to treat seeds were found.Thus, the purpose of the present work was to evaluate physiological (germination and vigor) and sanitary (incidence of fungi) quality in seeds of different lots of melon seeds treated with five doses of Thiabendazole.

Seeds of three lots of Cozumel hybrid melon were analyzed concerning physiological and sanitary qualities after treatment with the fungicide Tecto^® SC. This product belongs to the chemical group of Benzimidazole and the active ingredient (a.i.) is Thiabendazole (485 g/L of Thiabendazole). The experiments were conducted independently (one for sanitary and another for physiological quality), under a completely randomized design, with 15 treatments, according to a 3x5 factorial system (three seed lots: 67594, 84538 and 84539 × five fungicide doses: 0.0; 0.12; 0.2; 0.3 and 0.4% a.i.).

 

The application of fungicide was done in rotating pans with a central disk inside the pan located in the center, also rotating, but in the opposite direction to distribute the product. After treatment and drying, the seeds were evaluated for physiological quality and incidence of fungi.

 

Seed germination

 

Standard Germination Test (SGT) was done according to Seed Analysis Standards (ISTA, 2004; Brasil, 2009). The seeds were distributed over two pieces of paper towel dampened with 2.5 times their weight of distilled water and covered with another piece of moistened paper towel. The sheets were rolled and placed in a germination chamber in a vertical position at 25±1°C, relative humidity between 80 to 95%, without light. There were four replications with 50 seeds, amounting to 200 seeds per treatment. The counting of normal seedlings was performed on the 8^th day after sowing (DAS), with values expressed in percentages.

 

Seed vigor

 

Germination First Count (GFC) tests: normal seedlings were counted on the 4^th DAS during the SGT, as described previously, according to Brasil (2009), with values expressed in percentages.

 

Seed pathology analysis

 

The "blotter test" method was used to evaluate the incidence of fungi in seeds. It involved distributing ten seeds placed equidistantly over three sheets of filter paper, previously moistened, positioned on Petri dishes. Eight replications of 50 seeds were performed, totaling 400 seeds per treatments. The experimental plot consisted of five dishes with ten seeds.

 

The dishes were kept at 20±2°C for a twelve-hour photoperiod under white fluorescent light for seven days. After incubation, the seeds were individually evaluated under a magnifier and the results were expressed in percentage of seeds with fungus. The results were converted in arc sin Ö(x/100) for statistical analysis. The data obtained for all variables evaluated were subjected to variance analysis and the averages were compared against the Tukey test at 5% probability.

There was no interaction between factors (lots and fungicide doses), indicating mutual independence for germination and vigor. The results for both variables are shown in Table 1.

 

 

Although, all three melon seed lots offer excellent quality with minimum germination rate above 90%, lot 84538 showed lesser germination (90.8%) compared to the other two lots (67594 and 84539) (Table 1). The rates demonstrated for germination on three lots are much higher than the minimum standard rate of 70% for melon required for marketing in Brazil by the Ministry of Agriculture, Livestock and Supply (MAPA).

 

However, according to Cardoso et al. (2014), due to higher cost of hybrid seeds and competition between companies for the market, the official standards are outdated in relation to market practices. It is rare to find Cucurbitaceas seeds in the market with less than 85% germination rate. Even with this reference frame, all the lots can be considered to have good physiological quality.

 

The first seed count is considered a vigor test. The samples with faster germination and higher percentage of normal seedlings on this date are considered more vigorous (Marcos Filho, 2005; Baalbaki et al., 2009). There was no difference between lots in  this  feature (Table 1), with an average of 93.2% rate, indicating high seed vigor. Considering Thiabendazole dosage (Table 1), there was no difference in germination (average of 94.8%) and vigor (first count, average of 93.2%), showing that the fungicide did not affect physiological quality of melon seeds in any lot. According to Groot et al. (2006, 2008), Lobo (2008) and Menten and Moraes (2010), sensitivity to fungicide treatment can vary according to the initial quality of seed lots. In this study, despite differences in germination among lots, none of the seeds in any of the lots were affected by the fungicide treatment, regardless of the dose.

 

Garcia Júnior et al. (2008) found no phytotoxic effect of this fungicide on wheat seeds, verifying germination and seedling emergence similar to untreated control seeds, as observed by Mendes et al. (2001) with alfalfa seeds and Gally et al. (2004) with soybean seeds. Yet, Fessel et al. (2003) reported that a very high concentration of fungicide could impair maize seed quality. In contrast, Carvalho et al. (2004) found higher maize seed germination in different treated lots compared to the untreated control lot, because fungicide reduced the incidence of "damping off" caused by the fungus Stenocarpella maydis in seeds.

 

Also, Kaiser and Hannan (1987) verified an increase in seedling emergence for lentil seeds treated with Thiabendazole, mainly due to control of fungus Ascochita lentis. But, in high doses, this fungicide caused phytotoxic effects on lentil seedlings. The positive effects on germination and/or seedling emergence with seed treatment using this fungicide were also described in papaya seeds (Campos et al., 2009) and  castor  bean  seeds  (Poletine  et  al.,  2012).  These contrasting results for each crop show the importance of conducting studies for each different species because the results may not always be the same.

 

The treated seeds, regardless of dose and lot, showed no difference compared to untreated control seeds, both for germination test and vigor (Table 1). The discrepancy of reports on other species probably occurred for two reasons: (a) because the observed presence of fungi in seeds was small in all treatments, including in the control (Tables 2 and 3), (b) the treatment with this fungicide did not affect germination and vigor.

 

 

 

 

between factors (seed lots and Thiabendazole doses) was    significant (Table 3).  These    fungi     incidence decreased in seeds with increasing fungicide doses. Thus,   the   fungicide   showed  a  positive  effect  on  the control of these fungi species in melon seeds.

 

There have been no reports on melon seed treatment with this fungicide and related control of its pathogens. But, there have been reports on controlling Fusarium oxysporum in alfalfa seeds (Mendes et al., 2001), Aspergillus sp. and Penicillium sp. in peanut seeds (Barbosa et al., 2013), Ascochita lentis in lentil seeds (Kaiser and Hannan, 1987), Aspergillus niger, A. favus, Botrytis ricini, Curvularia sp., Penicillium sp. and Rhizopus sp. in castor bean seeds (David et al., 2014), Leptosphaeria maculans in cabbage seeds (Maude et al., 1984), Fusarium graminearum in wheat seeds (Garcia Júnior et al., 2008), Fusarium moniliforme (Pinto, 2000) and Stenocarpella maydis (Carvalho et al., 2004) in maize seeds. These studies demonstrate that Thiabendazole is effective in controlling fungi in seeds of different species, as observed in the present study with melon seeds.

In general, the fungicide Thiabendazole promotes a reduction of saprophytic fungus in melon seeds without impairing their germination and vigor.

The authors have not declared any conflict of interest.