Materials and methods

Study site. The study was conducted in the community of Acuitlapilco, Coatepec Harinas, Mexico (18º 54’ 20” N and 99º 47’ 12” W, at 2104 masl). Two independent experiments were established. The first was established from June 01-October 16, 2016, and the second from March 20-July 25, 2017, leaving 300 m between the two experiments. In order to analyze the soil characteristics for both experiments, a composite soil sample from five subsamples was used. Soil analyses were carried out at the Instituto de Investigación y Capacitación Agropecuaria Acuícola y Forestal (ICAMEX). The results indicated that the soil pH was moderately acid (6.5) and suitable for chrysanthemum production (García, 2014). The organic matter content was high (2.5 %); the nitrogen content (0.2 %) indicated that the soils are moderately rich in this element; the potassium content was 326-419 ppm; and the phosphorus content was 120-150 ppm (Rodríguez and Rodríguez, 2011), so there was no lack of these nutrients.

Biological material. For the experiment, the Chena, Flamingo and Moreliana cultivars were used, all of them from the hybrid complex Dendrathem grandiflora Tzevelev, and with a 90-120-day production cycle. Chena, with a white head and a cream-colored center, is the most widely planted cultivar in terms of area in the flower producing region of the southern part of the State of Mexico, but it is susceptible to B. cinerea. The Flamingo cultivar produces pink flowers, while the Moreliana cultivar produces white flowers with a purple center; both have similar market demand and are reported to be resistant to B. cinerea (Figure 1).

Experimental design. The Chena (Ch), Flamingo (F) and Moreliana (M) cultivars were planted in microtunnels, on 9 beds 50 m long and with 10-cm spacing between lines and between plants. The experimental design included blocks with randomly distributed plots. Three spatial arrangements (A) were established: A1=F-Ch-F; A2=M-Ch-M and A3=Ch-Ch-Ch, each with three repetitions. Arrangements A1 and A2 had 16 lines of the Chena cultivar in the middle flanked by four lines of a resistant cultivar (Flamingo or Moreliana) at both ends of the microtunnel, where the fungus is more frequently found due to water accumulation caused by runoff from the plastic cover. In arrangement A3, all the lines (24) were sown with the Chena cultivar, the same way flower growers do.

Figure 1 Chrysanthemum cultivars (Dendrathema grandiflora Tzevelev). (A) Chena, (B) Flamingo, (C) Moreliana. 

Agricultural and agronomic practices. Safe plants were provided by a cutting producer from the community of Zacango, Villa Guerrero, Mexico. Initially drip irrigation was applied up to 30 days after sowing, and furrow irrigation was applied after that. Eight days after sowing, the main stem was trimmed in order to promote side shoot growth. Then, the first NPK (18-46-00) fertilizer was applied. The crop was kept under yellow artificial light 24 h a day during 20 days after pruning, and the light was distributed in three lines 1.5 m high. Manual weeding and hoeing were carried out during the crop cycle. When the plants were approximately 50 cm high, raffia twine was wrapped around them. When flower buds emerged, the side buds were removed from each stem and only the middle bud was left. Commercial agrochemicals Lucaphos^® (diclorvos) (800 mL ha^-1) were used to control trips (Frankliniella occidentalis Pergande), Abamectina^® (avermectin B1a and avermectine B1b) (0.5 L ha^-1) for red spider (Tretanychus urticae Koch) and Beleaf^® (pridinacarboxamide) (200 g ha^-1) for whitefly (Trialeurodes vaporariorum Westwood). All these products were applied once a week during the production cycle, and Mancozeb® (manganese ethylene-bis-dithiocarbamate along with zinc ion) was used to prevent white rust (Puccinia horiana Hennings) (1 kg ha^-1) and B. cinerea, at the crop’s initial stage. Conventional chrysanthemum production requires at least four fungicide applications a week, and more combined products, such as Zineb^®, Manzate^®, Ziran^®, can also be used (personal communication with flower producers).

Evaluated variables. Three weeks after transplanting, the plants were measured every seven days until the cutting season. Seven readings of the plant height and stem thickness variables were taken on samples of 24 plants of the Chena cultivar randomly selected from each arrangement. The presence of the fungus was monitored during crop development according to the 1-9 scale described by ^{Abawi and Pastor-Corrales (1990}), where 1 = no visible symptoms, 5 = up to 25 % of foliar tissues show necrotic lesions characteristic of B. cinerea (Piesik et al., 2015), and 9 = 75 % or higher percentage of damaged leaves and dead plants. In the case of the head, we looked for brown spots on the ligules. During cutting, the diameter of the flower head was measured, and the vase life was evaluated for 16 days using 20 flower stems of Chena randomly selected from each spatial arrangement. The stems were placed in water in a cool, shady place to measure the percent infection on stems whose ligules and leaves showed brown spots caused by B. cinerea. The economic analysis was performed based on the price of the agrochemicals used, area sown (1125 m²), number of plants of the Chena cultivar (12,000 in A3 and 8,000 in A1 and A2), Chena cultivar cost (around 20 % higher than that of the Flamingo and Moreliana cultivars), as well as the number of additional Moreliana flowers, which unlike the Chena cultivar, keep their side stems. The statistical analysis of the plant height, the stem thickness, the floral diameter and the postharvest life variables was performed using analysis of variance (ANOVA) and the SAS System version 70 for Windows. The difference among the strategic arrangements set out for each variable was determined using the minimum significant difference test (MSD, P≤0.05).

Results and discussion

Stem thickness. There were significant differences (P≤0.05) in stem thickness among arrangements, where the diameter ranged from 0.62 to 0.72 cm (Table 1). The Chena cultivar in the traditional arrangement (A3) had the lowest values in all measurements compared to Flamingo (A1) and Moreliana (A2). In arrangements A1 and A2, the Chena cultivar had an advantage in one of the most important characteristics of the crop’s quality (Flores-Ruvalcaba et al., 2005), where stem thickness determines the weight of the flower head that the stem can support. The stem thickness is not affected by B. cinerea because the fungus infects old and soft tissues (Palmieri and Dardón, 2012; Schumacher, 2017). However, when A1 and A2 were intercropped, plants of the Chena cultivar were more vigorous, a fact that, along with the absence of B. cinerea, provides commercial quality advantages, as mentioned by ^{Costa et al. (2019}), who suggest that the greater height can be attributed to morphological, physiological and/or competition responses, characteristics that were not considered in this study.

Plant height. In arrangement A3, with significant differences (P≤0.05), Chena’s height (64.5 cm) was lower than the height of the same variety in arrangements A1 (70.6 cm) and A2 (69.9 cm), which were combinations of Chena and Moreliana or Flamingo (Table 2). Spatial arrangements usually vary in population density because of the distance between rows and plants, a fact that influences not only the yield components but also the products’ quality (Jirmanová et al., 2016; ^{Coelho et al., 2016}). However, in this study, the sowing density was the same in the three arrangements, which suggests that due to possible variations in nutrient demands (Valdez-Aguilar et al., 2015) caused by the strategic distribution of the cultivars, Chena plants were taller in A1 and A2 compared with those in A3. Piesik et al. (2015) mentioned that the association of certain chrysanthemum cultivars in the greenhouse is decisive to protect some of them. The plant height variable is also very important because in the marketing chain, taller stems make it possible to make cuts at their base without affecting quality (Flores-Ruvalcaba et al., 2005). Similarly, taller plants have competitive advantages, such as greater light gathering and faster nutrient assimilation (Zheng and Van Labeke, 2018). The flower head is more exposed to sunlight and avoids the higher relative humidity present in the plant’s low or intermediate parts, a factor that determines spore germination and mycelium growth (Palmieri and Dardón 2012; Schumacher, 2017). In this regard, although the plant’s height was not directly affected by the presence of B. cinerea, it could be an advantage that limits the damage to the flower head.

Flower head diameter. The largest diameter of a flower head and with significant differences (P≤0.05) was produced by the Chena cultivar in arrangement A2, followed by arrangements A1 and A3 (Table 3). The largest diameter of a flower head in A2 and A1 represented a commercial advantage because it is one of the most important characteristics of chrysanthemum quality (Flores-Ruvalcaba et al., 2005). This indicated that in the monocropped arrangement that included the Chena cultivar (A3), the flower head had no commercial advantage.

Vase life. Measurement of vase life was interrupted on day 16, when the Chena plants in A3, with significant differences (P≤0.05), were 100 % infected by B. cinerea (Table 4). However, according to estimations, this species may have a vase life of up to 30 days due to its low production of ethylene and because it is considered a non-climateric species (Hidalgo et al., 2011).

The presence of B. cinerea on the chrysanthemum crop was not detected on the foliage during crop development, but when randomly selected plants from arrangement A3 (conventional crop) were cut, we noted fungus growth (Table 4); 25 % of the plants had visible symptoms on the flower head, including brown spots and ligule dehydration. Just as in other flowers, such as rose (Rosa x hybrid), B. cinerea infection is not visible when flowers are cut, but it is latent and appears under moisture conditions during storage and transportation (Elad, 1988). Chena plants in A2 did not show infection by B. cinerea during the vase life period we evaluated, which indicates there is a commercial advantage when using this arrangement for this cultivar. These results show how important it is to control this fungus when the crop starts to develop.

The obtained results have an economic impact because a conventional arrangement to produce the Chena cultivar involves the use of fungicides such as Mancozeb, which must be applied four times a week during the rainy season, at an estimated cost of MX$2540.00 ha^-1 (US$133). Using arrangements A1 and A2, we were able to control B. cinerea with a single application at a cost of MX$645.00 ha^-1 (34 USD), which represented savings of 75 %. On the other hand, although there is a price difference, 20 % more in Chena compared with Moreliana and Flamingo, and a lower number of them per square meter in A1 and A2 compared with A3, earnings are compensated because one Moreliana stem produces 5 flowers on average.

The proposed spatial arrangement allows (Flamingo or Moreliana) cultivars located on the sides (where there is usually runoff of rain water or water that condenses during early morning hours) may act as a physical barrier, which in turn prevents moisture that favors fungus development on the Chena cultivar, which is more susceptible to B. cinerea infection. Other studies (García-Velasco et al., 2003; Solano-Baez et al., 2013) have shown that there are chrysanthemum cultivars that are more resistant to fungal damage, so it is important to understand the type of physiological response, including the release of volatile organic compounds that inhibit pathogens (Piesik et al., 2015), as well as to identify cultivars with potential resistance to pathogens that could be used as physical barriers in spatial arrangements such as the one used in this study or in improvement programs.

Conclusions

B. cinerea incidence was significantly lower in plants close to the Chena cultivar during cutting and in terms of vase life, when it was flanked by Flamingo (A1) or Moreliana (A2). Similarly, the commercial quality of the Chena cultivar’s floral stems was better in spatial arrangements A1 and A2 compared with those in A3, in which Chena was monocropped.