The greenhouse tomato production system commonly practiced in Northern Europe and North America, including Mexico, is based on the use of varieties with an undetermined habit (Sánchez-del Castillo, Bastida-Cañada, Moreno-Pérez, Contreras-Magaña, & Sahagún-Castellanos, 2014). In this system, seedlings are transplanted into improved soil or hydroponic substrates at densities of 2 to 3 plants·m-2, and are tutored with one or two stems up to 3 m in height to harvest 15 to 20 clusters per plant in a 10 to 11 month crop cycle, and 7 to 8 month harvest period (Sánchez-del Castillo et al., 2014). Under these conditions, yields of 400 t·ha-1·yr-1 can be achieved, although production costs are very high (Ramos-Fernández, Ayala-Garay, Pérez-Grajales, Sánchez-del Castillo, & Magdaleno-Villar, 2021).
As an alternative, a production system based on establishing several crop cycles in one year has been developed (Ponce-Ocampo, Sánchez-del Castillo, Contreras-Magaña, & Corona-Sáez, 2000). This is achieved by plant blunting (in which the growth apex and lateral shoots are removed) in a production system with few clusters (1 to 5 per plant) and high population densities (8, 9, 12, 16, 20 or more plants·m-2) (Sánchez-del Castillo, Moreno-Pérez, Vázquez-Rodríguez, & González-Núñez, 2017; Villegas-Cota et al., 2004; Santos & Sánchez-del Castillo, 2003). Blunting decreases plant height from 2-3 m to only 90-100 cm; with this, the period from transplanting to the end of harvest is reduced by 66 %, from 270 to 90 days approximately, with a 180 t·ha-1 yield per cycle (Sánchez-del Castillo, Moreno-Pérez, & Contreras-Magaña, 2012). By reducing the crop cycle, phytosanitary problems can be reduced and the harvest time shortened, and this can be used for programming the harvest to occur in good-price market windows and improving the producer's income. Likewise, lower greenhouses with a lesser degree of technification can be used, and this, in turn, can reduce production costs per cycle (Sánchez-del Castillo et al., 2017).
Under this system, it is possible to obtain three annual production cycles, which is achieved by maintaining the crop for 30 days in the seedbed and 105 days in the field. However, the plant uses the first 60 days only for vegetative growth. In this regard, in order to make the process more efficient, the plant has to develop those 60 days in the seedbed, so that production in the field would only take 75 days (2.5 months). In this way, two management options are possible: one with four annual cycles with three clusters and the other with five clusters per plant and three cycles per year (90 days per cycle and 8 plants·m-2).
On the other hand, keeping the seedlings in containers for 60 days may result in plants with thin leaves and stems, increased height and presence of flowers, which describes a poor quality seedling. This is due to competition among plants for light, nutrients and space for root development. An alternative to avoid poor quality plants is the use of growth retardants, which are substances that inhibit gibberellin synthesis and shorten internodes, which reduces plant height and their leaf area (Jankiewicz, 2003; Rademacher, 2000; Ramírez et al., 2005; Ugur & Kavak, 2007).
Growth retardants are used to obtain low, compact plants, with thick stems and vigorous leaves, traits desired in the production of high quality tomato seedlings. Therefore, the aim of this research was to produce tomato plants in a seedbed with sufficient quality for intensive greenhouse cultivation, using plant growth retardants.
Materials and methods
This study was carried out during the autumn-winter cycle in a chapel-type greenhouse with a metallic structure and polyethylene cover (75 % light transmission), located at 19º 20’ N, 98º 53’ W and 2,250 m a. s. l. in Chapingo, State of Mexico.
Sowing was carried out in 60-cavity polystyrene trays, with each cavity having a volume of 170 mL. The substrate used was a mixture of peat moss and perlite at a ratio of 2:1 (v/v). Two seeds were deposited per cavity; once they emerged, one seedling was removed so that all cavities would have only one before treatment application. The DRD 8537 hybrid known as “Martino 37” (Seminis, St. Louis Missouri, USA) was used as plant material; this has a ball-type fruit and a determinate growth habit, and can form up to five clusters.
Three growth retardants were tested: Uniconazole (Sumagic®, Valent, Chile), Paclobutrazol (Cultar® 25 SC, Syngenta, Switzerland) and Propiconazole (Tilt®, Syngenta, Switzerland), with two application levels (first level: one application at 25 days after sowing [das], second level: applications at 25 and 50 das), and two application doses of each product (low and high; factor nested in products since the doses were different in each case). In addition, an absolute control without applications was used, resulting in 13 treatments (Table 1).
|Treatment||Retardant||Active ingredient||Application period (das)||AI doses (mg·L-1)|
|7||Cultar 25sc®||Paclobutrazol||25, 50||25|
|8||Cultar 25sc®||Paclobutrazol||25, 50||50|
|10||Tilt 250sc ®||Propiconazole||25||100|
|11||Tilt 250sc®||Propiconazole||25, 50||50|
|12||Tilt 250sc ®||Propiconazole||25, 50||100|
|13||No application||No application||No application||No application|
A completely randomized design with a factorial arrangement (3x2x2) and four replicates was used. The experimental unit (EU) was 15 seedlings, so a complete tray was used for the four replicates. The retardants were applied by foliar sprays until completely covering the foliage of the plants, with the same number of sprayings in each case. Solution doses of 10 and 18 mL were used per container in the first and second applications, respectively. Manual atomizers of 1 L capacity were used for spraying, one for each product.
Five plants were taken from the center of the EU per treatment in order to quantify the response variables.
Plant height (PH, cm). It was measured from the stem’s base to the growth apex at 31, 49 and 66 das.
Stem diameter (SD, mm). It was measured with a digital Vernier (model 14388, Truper®, China) 1 cm above the cotyledon leaves at 31, 49 and 66 das.
Number of leaves (NL). Leaves present at 31, 49 and 66 das were counted. Only well-formed leaves were considered, that is, those with leaf lamina and visible leaflets.
Leaf area (LA, cm2). Leaves with the petiole were placed in a leaf area meter (LI 3100, Li-Cor®, USA). This variable was determined only at 66 das, by destructive sampling.
Fresh weight (FW, g). At 66 das, five plants were taken from each EU, the substrate was removed with water, the excess water was drained off and the plants were weighed on a balance (Pioneer, Ohaus®, USA) with 0.01 g accuracy.
Dry matter (DM, g). The same five plants used for FW were placed in paper bags and taken to a drying oven (FE-291, Felisa®, Mexico) where they remained at 65 °C until constant weight. Afterwards, they were weighed on the abovementioned balance.
An analysis of variance was carried out with the data obtained in each sampling (31, 49 and 66 das). Additionally, Tukey's comparison of means (P ≤ 0.05) was carried out in two ways: a) with a factorial arrangement without considering the control to identify the best retardant product, dose and number of applications, and b) with a completely randomized design that included the absolute control, in order to obtain the best treatment or treatments with respect to the control. The SAS statistical package version 9.1.3 (SAS Institute, 2003) was used for this purpose.
Results and discussion
The statistical analysis showed that at 31 das there was a significant difference (P ≤ 0.05) in SD and a highly significant difference (P ≤ 0.01) in PH, while no differences were observed in NL. In the following evaluations (at 49 and 66 das), all these variables presented highly significant differences, except for NL at 66 das, which was only significant (Table 2).
|SV||PH (cm)||NL||SD (mm)||PH (cm)||NL||SD (mm)||PH (cm)||NL||SD (mm)||FW (g)||LA (cm2)||DM (g)|
|31 das||49 das||66 das|
The comparison of means with factorial arrangement indicated that, for the retardant factor, Paclobutrazol had lower values (P ≤ 0.05) than Propiconazole in the variables evaluated at three times (31, 49 and 66 das), except for NL, where values were similar (Table 3). In the number of applications factor, a single application was sufficient to retard seedling development for two months; only PH was affected at 31 and 49 das, although in both cases only the first application had been carried out. No significant differences were found at 66 das in the variables evaluated, except for NL and SD, although NL was the same at the end of the seedbed period (seven leaves per plant), whereas for SD, two applications allowed this parameter to be significantly higher (Table 3).
|Factor||PH (cm)||NL||SD (mm)||PH (cm)||NL||SD (mm)||PH (cm)||NL||SD (mm)||FW (g)||LA (cm-2)||DM (g)|
|31 das||49 das||66 das|
|Uniconazole||6.06 bz||2.00 a||3.41 a||15.53 b||4.00 b||5.00 b||20.21 b||6.88 a||5.10 b||19.48 b||143.67 b||2.76 b|
|Paclobutrazol||6.16 b||2.00 a||2.66 b||13.75 c||3.94 b||4.80 b||17.73 c||6.50 a||5.20 b||16.72 c||117.89 c||2.26 c|
|Propiconazole||7.08 a||2.00 a||3.33 a||21.49 a||4.75 a||5.80 a||26.11 a||6.81 a||5.56 a||24.73 a||218.71 a||3.17 a|
|One||6.19 b||2.00 a||2.97 a||16.28 b||4.21 a||5.26 a||21.48 a||6.88 a||5.20 b||20.45 a||163.93 a||2.77 a|
|Two||6.68 a||2.00 a||3.30 a||17.57 a||4.25 a||5.15 a||21.22 a||6.58 b||5.37 a||20.17 a||156.25 a||2.69 a|
|Low||6.74 a||2.00 a||3.06 a||17.37 a||4.25 a||5.24 a||22.05 a||6.62 a||5.26 a||20.57a||170.02 a||2.75 a|
|High||6.12 b||2.00 a||3.20 a||16.47 b||4.20 a||5.16 a||20.64 b||6.83 a||5.30a||20.04 a||150.14 b||2.70 a|
In the dose factor, significant differences (P ≤ 0.05) were found in PH and LA; in both cases, the high dose reduced them. The variables NL, SD, FW and DM were not affected by doses applied at 66 das (Table 3).
According to the factorial mean test, when analyzing the effect of the retardants at the three times evaluated, it was observed that the seedlings with the lowest height at 31 das were those sprayed with Uniconazole, although they were statistically equal to those sprayed with Paclobutrazol. In contrast, Paclobutrazol was the product that decreased PH the most at 49 and 66 das, while Propiconazole affected height the least on the three dates evaluated (Table 3). According to Zandstra, Dick and Lang (2004), within the triazol group, Propiconazole is less active than Uniconazole and Paclobutrazol, which coincides with what was observed in this study. The physiological effect of triazoles in reducing PH is due to the inhibition of gibberellin synthesis (Jankiewicz, 2003), which has been observed in several plants: Paclobutrazol in tomato (Ramos-Fernández et al., 2021), eggplant, bell pepper (Partida-Ruvalcaba et al., 2007), chili (Velázquez-Alcaraz, Partida-Ruvalcaba, Acosta-Villegas, & Ayala-Tafoya, 2008), and arabidopsis and maize (Hartwig et al., 2012), as well as Paclobutrazol and Uniconazole in potato (Flores-López et al., 2011; Flores-López, Martínez-Gutiérrez, López-Delgado, & Marín-Casimiro, 2016) and kalanchoe (Currey & Erwin, 2012).
When comparing means at 66 das, with the inclusion of the control (Table 4), it was observed that the plants treated with a growth retardant were shorter than the control plants (T13, without application). The maximum reduction in height was achieved when 50 mg·L-1 Paclobutrazol was applied at 25 das (T6). In this case, the height reduction was 13.15 cm; that is, 44% less than the control treatment.
|Treatment, active ingredient, period, doses (mg·L-1)||PH (cm)||NL||SD (mm)||FW (g)||LA (cm2)||DM (g)|
|T1, UCZ, 25, 2.5||19.60 defz||7.00 ab||5.09 defg||18.23 de||131.11 fgh||2.68 bc|
|T2, UCZ, 25, 3.0||19.95 de||7.00 ab||4.90 g||18.87 d||137.50 fg||2.72 bc|
|T3, UCZ, 25 and 50, 2.5||22.04 cd||6.50 bc||5.41 bc||21.97 c||168.58 de||3.10 a|
|T4, UCZ, 25 and 50, 3.0||19.27 efg||7.00 ab||4.98 fg||18.87 d||137.50 fg||2.55 cd|
|T5, PBZ, 25, 25||19.16 efg||6.50 bc||5.21 cdef||18.73 d||144.26 ef||2.42 cd|
|T6, PBZ, 25, 50||16.88 g||7.00 ab||5.01 efg||15.51 f||98.72 i||2.24 de|
|T7, PBZ, 25 and 50, 25||17.32 fg||6.00 c||5.16 cdefg||15.61 f||117.20 ghi||1.94 e|
|T8, PBZ, 25 and 50, 50||17.56 efg||6.50 bc||5.42 bc||17.04 def||111.37 hi||2.42 cd|
|T9, PCZ, 25, 50||26.82 b||7.25 a||5.35 bcd||25.01 ab||245.23 a||3.23 a|
|T10, PCZ, 25, 100||26.50 b||6.50 bc||5.65 ab||26.37 a||226.76 ab||3.31 a|
|T11, PCZ, 25 and 50, 50||27.40 b||6.50 bc||5.37 bcd||23.92 abc||213.79 bc||3.14 a|
|T12, PCZ, 25 and 50, 100||23.74 c||7.00 ab||5.89 a||23.62 bc||189.05 cd||2.99 ab|
|T13, Control||30.03 a||7.00 ab||5.30 cde||23.22 bc||216.24 b||3.17 a|
The treatment that least affected PH was 50 mg·L-1 (T11) Propiconazole applied at 25 and 50 das, which resulted in a height 8.76 % lower than the plants without application, although it statistically exceeded the control (Table 4). This agrees with Zandstra et al. (2004), who observed that Propiconazole is less active than Paclobutrazol and should be used at higher doses. Berova and Zlatev (2000) reported a 16-20 % reduction in tomato seedling height with a 25 mg·L-1 Paclobutrazol foliar application. Similarly, Campos-de Melo, Seleguini, and Santos-Veloso (2014, 2015) observed a 24 and 25 % reduction in tomato seedling height when seeds were treated with 115.4 and 50 mg·L-1 Paclobutrazol, respectively.
If the ideal height is considered to be 20 to 25 cm, since plants are compact when they are two months old, treatments with Paclobutrazol and Uniconazole (T1 to T8) and Propiconazole with two applications and a high dose (T12) are the most appropriate. The other Propiconazole treatments (T9 to T11) did not reduce height to less than 25 cm.
Number of leaves
The use of retardants can affect the emergence period of leaves, but not their number, as occurred at 49 das (Table 3); this can be explained by the fact that this trait is defined by the genotype. Jankiewicz (2003) reports that growth retardants have no influence on the NL of treated plants.
According to the mean comparison test, at 66 das the plants with the lowest NL (6 leaves) were those treated with 25 mg·L-1 Paclobutrazol applied at 25 and 50 das (T7), which implies a 14.28 % reduction with respect to the control. The rest of the treatments showed the same NH as the treatment without application (Table 4). In this regard, Flores-López et al. (2011) found a reduction in NL (3 and 4) in potato 45 days after applying 20 and 40 mg·L-1 Uniconazole, and when they applied 200 and 250 mg·L-1 Paclobutrazol they observed a reduction of 3 and 6 leaves, respectively.
The NL in a plant is very important, since leaves, being the photosynthetic organs, are the main light receptors, so their number and size influence the efficiency of sunlight capture. This influences the overall development of the plant and, as a result, its yield, hence the importance of the number of leaves present (Reis, de Azevedo, Albuquerque, & Junior, 2013). However, in this study the use of growth retardants had no significant effect on NL, except for treatment T7 (Table 4).
Growth retardants affected SD. Seedlings treated with Propiconazole had a larger SD than those sprayed with Paclobutrazol on the three sampling dates. On the other hand, the performance of plants treated with Uniconazole varied among sampling dates (Table 3). Jankiewicz (2003) states that growth retardants induce the formation of thicker stems compared to untreated plants. Ferreira et al. (2017) found that tomato SD increased 17 and 11 % at 15 and 30 days after spraying the foliage with 42.5 mg·L-1 Paclobutrazol 10 das.
In the mean comparison test, considering the control, it was observed that plants that received two applications of Propiconazole (at 25 and 50 das), with doses of 100 mg·L-1 (T12), had the largest SD (5.89 mm); however, these plants did not differ statistically from those treated with only one application of the same product and dose (T10; 5.65 mm), which indicates that one application may be sufficient. In both cases, SD was higher than that of the plants with no application (5.30 mm). On the other hand, plants treated with 3 mg·L-1 Uniconazole at 25 das (T2), as well as those that received two applications of Uniconazole with the same dose (T4), had the smallest SD (4.90 and 4.98 cm, respectively), statistically lower than the control (T13) (Table 4). Ferreira et al. (2017) found that tomato SD increased 17 % after transplanting in plants treated with 42.5 mg·L-1 Paclobutrazol applied 10 das; this is an additional trait conferred by the use of triazoles (Jankiewicz, 2003).
In general, with the exception of the two Uniconazole treatments mentioned above, SD reached more than 5 mm, which is appropriate for the tomato seedling. This is because the greater the diameter and the lower the height, the possibility that the plant will flatten or bend after transplanting is reduced. Therefore, although the control had more than 5 mm in diameter, the height it developed was not adequate for transplanting. Similarly, Berova and Zlatev (2000) observed that foliar application of Paclobutrazol, at a 25 mg·L-1 dose, caused an increase in stem thickness and root development, improved photosynthetic activity and water balance (and thus the quality of seedlings for transplanting), and accelerated the formation and harvest of tomato (Lycopersicon esculentum Mill.) cv Precador, without leaving Paclobutrazol residues in the fruit.
Fresh weight, leaf area, and dry matter
The retardants affected FW, LA and DM of seedlings at 66 das (Table 3), with Paclobutrazol achieving the greatest reduction in all three variables (P ≤ 0.05), followed by Uniconazole and Propiconazole. This shows that Propiconazole is the least active (Zandstra et al., 2004). Flores-López et al. (2011) observed a 32 and 46 % reduction in leaf area index of potato at 45 days after applying 40 and 250 mg·L-1 of Uniconazole and Paclobutrazol, respectively, which allowed inferring the greater effect of Paclobutrazol.
When considering the control (Table 4), the comparison of means indicated that the highest FW was achieved with Propiconazole in a single 100 mg·L-1 application (T10) and two 50 mg·L-1 applications (T9) (at 25 and 50 das), with values of 26.37 and 25.01 g, respectively. These values exceeded the control by 14 and 8 %. On the other hand, the lowest FW was obtained with a 50 mg·L-1 application of Paclobutrazol at 25 das (T6), and with two 25 mg·L-1 applications (at 25 and 50 das) (T7); that is, both had 33 % less weight than the control. Treatments based on Uniconazole, in any of its doses and application times, also decreased FW at 66 das by 5 to 21 % compared to the control. In this regard, Partida-Ruvalcaba et al. (2007) observed that the application of 150 mg·L-1 Paclobutrazol increased the FW of the aerial part and root of bell pepper and eggplant.
Regarding LA, the mean comparison test (Table 4) indicated that the treatment with an application of Paclobutrazol at 25 das, with a dose of 50 mg·L-1 (T6), was statistically inferior to the control by 55 %. In general, the treatments with Paclobutrazol and Uniconazole reduced LA by between 22 and 55 % compared to the treatment without application, contrary to what was observed with Propiconazole, with which values equal to or greater than the control were obtained; that is, with this retardant no effect on LA was observed, with the exception of T12, which reduced it by 12 %. Flores-López et al. (2011) report that foliar applications of 250 and 40 mg·L-1 Paclobutrazol and Uniconazole decreased the LA index by 46 and 32 %, respectively, in potato 45 days after application.
The highest DM content was obtained with Propiconazole-based treatments, which were similar to the control and to the 2.5 mg·L-1 application of Uniconazole at 25 and 50 das (T3). On the other hand, the lowest DM content was obtained with the spraying of 25 mg·L-1 Paclobutrazol at 25 and 50 das and 50 mg·L-1 at 25 das (T7 and T6, respectively). It is important to emphasize that this variable is positively correlated with PH, FW and LA (data not shown), implying that the effect of retardants is similar in all cases. Similar results were reported by Nascimento, Salvalagio and Silva (2003) in tomato seedlings, where 200 mg·L-1 Paclobutrazol decreased the height and dry weight of the aerial part and roots, in relation to the treatments with Ethephon and without application. In contrast, Partida-Ruvalcaba et al. (2007) observed an increase in the FW of the aerial part and root of bell pepper and eggplant with the application of 150 mg·L-1 Paclobutrazol.
In general, it was observed that Paclobutrazol was the retardant that had the best response according to the objectives set and the type of seedling desired, due to the reduction in PH and LA, which translates into more compact seedlings. The use of Uniconazole could be a second option, and Propiconazole, with higher doses, could also be used, since it increased SD and induced a less intense green coloration and a lower degree of leaf curl than the other products (traits not evaluated), this in spite of not having greatly reduced the PH and LA of the seedlings with respect to the control.
The use of growth retardants (Uniconazole, Paclobutrazol and Propiconazole) had a significant effect on height, stem diameter, number of leaves, leaf area, fresh weight and dry matter in tomato seedlings. The number of applications did not significantly impact seedling traits, so one application is sufficient to induce the effect of retardants at this phenological stage. High doses caused a greater response in the variables evaluated. Paclobutrazol generated tomato seedlings with the best traits for use in late transplanting, followed by Uniconazole and Propiconazole.