Mexico produces 3.5 million tons of tomato (Solanum lycopersicum L.), of which 50 % are exported to the United States, generating 1.345 billion dollars annually. It is estimated that more than 50 % of what is exported comes from greenhouses and shade-houses (Servicio de Información Agroalimentaria y Pesquera [SIAP], 2017).
The predominant greenhouse tomato cultivation system in Mexico is based on that used in northern European countries and Canada, where varieties with an indeterminate growth habit are used with population densities of 2.5 to 3 plants·m-2 and more than 20 clusters per plant are harvested in an annual growing cycle with yields that can exceed 500 t·ha-1·year-1. This is achieved by controlling the environmental conditions that can be obtained with their highly technified and expensive greenhouses (Cheiri, de Gelder, & Peet, 2018; Heuvelink, Li, & Dorais, 2018). However, the lower technological level used in our country, reflected in greater pest and disease problems, or in shorter growing cycles, makes it difficult to exceed 300 t·ha-1·year-1 (Castellanos & Borbón-Morales, 2009). Furthermore, due to the very long harvest period (five to seven months), production costs frequently exceed $5,000.00 MXN per ton and the selling price in domestic markets, such as supply centers, fluctuates greatly; for this reason, the economic benefit is usually limited due to the high percentage of producers with less than one hectare of greenhouses who do not have access to other markets, such as the export one (Sánchez-del Castillo & Moreno-Pérez, 2017).
An alternative tomato production system has been developed and validated. It involves transplanting seedlings 45 to 50 days after sowing (das), and then leading the plants to a single stem and removing the terminal bud two leaves above the third inflorescence, this to leave only three clusters per plant. With the above, the cycle from transplanting to end of harvest is shortened to 110 days, and in a continuous production scheme, three growing cycles per year can be obtained with higher annual productivity than in conventional systems (Sánchez-del Castillo, Moreno-Pérez, & Contreras-Magaña, 2012). Due to the concentration of the period from the start to end of the harvest (30 days), it can be programmed so that it is obtained when there are high price windows, so the producer can have a greater economic benefit and also recover his investment faster (Sánchez-del Castillo et al., 2012). Likewise, the lower yield per plant in a growing cycle is partially compensated by the establishment of high population densities, ranging from 7 to 8 plants·m2 of greenhouse (Sánchez-del Castillo, Moreno-Pérez, Vázquez-Rodríguez, & González-Núñez, 2017).
Under this production system, average yields of 16 kg·m-2 per 3.5-month cycle are achieved (with a potential of 500 t·ha-1·year-1), with production costs per kg similar to the conventional system, but with fewer health problems because of the short cycles (Sánchez-del Castillo et al., 2012; Sánchez-del Castillo et al., 2017). The annual yield per unit area offered by this production system could be increased by shortening the time lapse from transplant to the end of the harvest to less than three months to obtain four growing cycles per year instead of three. This could be achieved by managing the seedlings in the seedbed in order to prolong their transplanting until 60 das without negative effects on the subsequent yield. Yield could also be increased by promoting the formation of more flowers and fruits in each of the three inflorescences per plant.
In the early stages of seedling growth, applying a growth retardant such as paclobutrazol can form plants with short internodes (Brigard, Harkess, & Baldwin, 2006) and smaller leaves (Seleguini, de Araujo-Faria, Silva-Benett, Lacerda-Lemos, & Seno, 2013), making it easier to keep the seedlings longer in the seedbed and to transplant at an older age without subsequent negative consequences.
Regarding the increase in the number of flowers per inflorescence, Heuvelink et al. (2018) point out that while this trait has a genetic component, it is possible to reduce the number of abortions of the flower primordia initiated in each inflorescence, which would make more of them reach anthesis. This can be achieved by temporary modifications of environmental conditions (light, temperature, CO2, nutrition, etc.) or the management of the source-demand relationship (use of hormones or growth regulators such as retardants). With greater spacing between seedlings in the seedbed (lower population density), each seedling is expected to receive the incident photosynthetically active radiation more evenly and thus increase its rate of photoassimilate production, leaving more sugar available for the flower primordia and with it the possibility of producing more flowers per inflorescence (Heuvelink & Okello, 2018).
Based on the above, the present study was carried out with the aim of evaluating the effect of paclobutrazol applications and different population densities in a seedbed on morphological characteristics related to the quality of tomato seedlings at 60 das, and on the number of flowers and fruits per inflorescence of plants pruned to the third cluster.
Materials and methods
The present research was carried out in greenhouses at the Universidad Autónoma Chapingo experimental field, located in Texcoco, State of Mexico (19° 20’ NL and 98° 53’ WL, at 2,240 masl). A gable greenhouse with a thermal polyethylene cover with 85 % transmission and 55 % light diffusion was used in the seedbed phase. It had polyethylene curtains, anti-aphid mesh, and a wet wall system and extractors that allowed maintaining a temperature of 15 to 25 °C during the day and 10 to 16 °C at night, with relative humidity between 50 and 70 % most of the day. Management after transplantation was carried out in another greenhouse with similar characteristics.
The hybrid tomato 'El Cid F1', which is a Harris Moran brand, was used for the experiment. This saladette-type hybrid has an indeterminate growth habit, and was chosen because it is widely used by growers due to its size, firmness, color and long shelf life, which is reflected in better selling prices. The seeds were sown in 60-cavity polystyrene trays with a volume of 250 mL per cavity, and 5 cm separation between the center of each cavity. A mixture of peatmoss and perlite (50/50, v/v) was used as substrate. The seedlings were irrigated with a nutrient solution containing the following nutrient concentrations (mg·L-1): 200 N, 50 P, 200 K, 250 Ca, 50 Mg, 150 S, 2 Fe, 1 Mn, 0.5 Bo, 0.1 Cu and 0.1 Zn. For the first 15 das, the nutrient solution was applied at half its concentration; then, until the end of the harvest, the full concentration was used.
Sixteen treatments were established, resulting from the combination of two population densities (150 and 300 seedlings·m-2), seven paclobutrazol (B-[(4-chlorophenyl)methyl]-α-(1,1-dimethylethyl)-N-1,2,4-triazole-1-ethanol) (Latimer, 1992) application treatments and two controls without application, one for each density. The paclobutrazol application treatments were: 1) one application of 50 mg·L-1 of active ingredient at 20 das, 2) one application of 50 mg·L-1 of active ingredient at 30 das, 3) one application of 50 mg·L-1 of active ingredient at 40 das, 4) two applications of 25 mg·L-1 of active ingredient at 20 and 40 das, 5) two applications of 50 mg·L-1 of active ingredient at 20 and 40 das, 6) three applications of 25 mg·L-1 of active ingredient at 20, 30 and 40 das, and 7) three applications of 50 mg·L-1 of active ingredient at 20, 30 and 40 das.
A randomized complete block experimental design was used with a split-plot treatment arrangement with four replicates. In the large plots, the population densities were located and in the subplots the paclobutrazol application treatments, with 15 seedlings per experimental unit.
Variables measured in the seedbed at 60 das were: 1) seedling height (measured with a tape measure), 2) stem thickness in the internode between the fourth and fifth leaf (measured with an electronic Vernier caliper [Digimatic Caliper CD-6 CS, Mitutoyo, USA]), 3) leaf area per plant, in two representative plants of each treatment in each replicate (measured with a portable leaf area meter [LI-3000A, LI-COR, Nebraska, USA]) and 4) total dry weight (obtained by oven drying to constant weight the same two plants per treatment in each sampled replicate to obtain leaf area and leaf area ratio [grams of total dry matter per m2 of leaf] from the same two plants per treatment in each replicate sampled to obtain leaf area).
Transplanting was carried out at 60 das, for which three rows of plants were placed in beds 1 m wide x 25 cm deep filled with volcanic sand (red tezontle) with particles from 1 to 3 mm in diameter. The paths left between beds were 50 cm wide. Nine plants were used per experimental unit, established at a distance of 30 cm between plants and 33 cm between rows.
A tape-based irrigation system was installed with integrated drippers every 20 cm. Irrigation was carried out with a nutrient solution with the concentrations of nutrients indicated above. There were between three to five irrigations daily (1 L·m-2 in each irrigation) according to weather conditions and the age of the plants. As part of the management, a preventive program was established for the control of pests and diseases, plant tutoring, pruning of side shoots to leave a single stem per plant and removal of the growth apex two leaves above the third inflorescence formed. At this stage, the variables evaluated were: 1) number of flowers per inflorescence in the three clusters per plant, 2) number of fruits per plant, 3) average fruit weight and 4) total yield per plant.
Analysis of variance was applied to the data obtained, followed by Tukey’s comparison of means test (P ≤ 0.05).
Results and discussion
Seedling quality indicators at transplant
Analysis of variance (data not shown) indicated significant differences (P ≤ 0.01) for seedling population densities in the seedbed and paclobutrazol treatments in all seedling quality traits assessed at 60 das, except for leaf area per plant in the density factor. The population density x paclobutrazol interaction was not significant in any of the traits evaluated. Coefficients of variation were generally low, ranging from 4 to 12 %.
The comparison of means tests, in the average of the paclobutrazol application treatments (Table 1), show that the seedlings that grew at low population density (150 seedlings·m-2) had 4.3 cm less height, 0.23 mm more stem thickness, 1.44 g more dry weight and 20 cm2·g-1 less leaf area ratio, compared to those that grew at 300 seedlings·m-2. All differences were statistically significant, except for leaf area.
|Treatment||Height (cm)||Stem thickness (mm)||Leaf area (cm2)||Dry weight (g)||Leaf area ratio (cm2·g-1)|
|150 seedlings·m-2||33.7 bz||5.35 a||703 a||7.54 a||94 b|
|300 seedlings·m-2||38.0 a||5.12 b||690 a||6.10 b||114 a|
The results obtained are in line with those indicated by Taiz and Zeiger (2002) in the sense that, above a certain population density threshold, competition between plants for photosynthetically active radiation causes etiolation symptoms such as an increase in stem length and a reduction in their diameter. At the same time, the photosynthesis per plant decreases and thus its accumulated dry weight, so that at a higher density there is a lower production of assimilates per plant, which are used more in stem elongation than in cell division and growth to thicken tissues. In the present study, at 60 das the seedlings at low density (150 seedlings·m-2) reduced their height by 11.3 %, but increased their stem diameter by 4 % and their dry weight by 24 %, this compared to the seedlings that grew at the higher density (300 seedlings·m-2).
Stem thickness and seedling dry weight were greater with low seedling density, possibly because as the seedlings are further apart, they intercept radiation more efficiently and produce more photoassimilates per day (Soltani & Sinclair, 2012).
Giovinazzo and Souza-Machado (2001) found that, with low population density, the leaf area developed by each seedling was greater than with high density. This was not the case in the present experiment, possibly because with the greater spacing between seedlings the leaves intercepted the incident photosynthetically active radiation incident more evenly, resulting in the formation of more sugars that could favor the formation of thicker leaves. This can be indirectly inferred from the lower leaf area ratio in the low-density treatment, which in turn contributed to the higher total dry weight found.
In the average density (Table 2), the paclobutrazol applications significantly reduced seedling height with respect to the control. This effect was greater with three applications of 50 ppm of active ingredient, reducing the height by 9.4 cm relative to the control (14 % less). Brigard et al. (2006) and Seleguini et al. (2013) state that in the seedling stage with the application of paclobutrazol, plants with shorter internodes are formed, which makes it possible to keep the seedlings longer in the seedbed. It should be noted that paclobutrazol is absorbed by the aerial part and is translocated via xylem to the growth points where it inhibits the production of gibberellins by preventing the oxidation of kaurene to kareuranic acid, thereby reducing the rate of cell division and expansion, which limits growth (Rademacher, 2000).
|Treatment||Height (cm)||Stem thickness (mm)||Leaf area (cm2)||Dry weight (g)||Leaf area ratio (cm2·g-1)|
|Control without application||40.6 az||5.12 bc||826 a||7.67 a||110 abc|
|50 ppm, 20 das||36.0 cd||4.91 c||608 c||6.33 b||98 bc|
|50 ppm, 30 das||34.2 de||5.16 abc||663 bc||6.28 b||107 abc|
|50 ppm, 40 das||38.6 ab||5.50 a||789 a||6.88 ab||117 ab|
|25 ppm, 20 and 40 das||37.1 bc||5.30 ab||733 ab||6.21 b||119 a|
|50 ppm, 20 and 40 das||35.5 cde||5.17 abc||646 bc||7.08 ab||92 c|
|25 ppm, 20, 30 and 40 das||33.6 e||5.39 ab||678 bc||7.34 ab||95 c|
|50 ppm, 20, 30 and 40 das||31.2 f||5.31 ab||631 bc||6.80 ab||95 c|
Stem thickness was significantly reduced relative to the control in most paclobutrazol application treatments; only the late application (40 das) of 50 ppm resulted in thicker stems (P ≤ 0.05). Giovinazzo and Souza-Machado (2001), when applying paclobutrazol in tomato, report a 9 % increase in stem diameter. Sun, Chen, Chang, Tseng, and Wu (2010) also highlight greater stem vigor with the application of paclobutrazol to tomato seedlings. Possibly the discrepancies with the present work are due to differences in doses, varieties tested and experimental conditions.
Several of the treatments with paclobutrazol application caused a significant reduction in the leaf area of the seedling. The most noteworthy treatments were those of one application of 50 ppm at 20 das and those of three applications of 50 ppm at 20, 30 and 40 das, which reduced this variable by 218 and 195 cm2 per seedling, respectively, with respect to the control without application (equivalent to a decrease of 26 and 24 %, respectively). On the other hand, the triple application treatments of paclobutrazol at any of its doses (25 or 50 ppm) maintained a similar dry weight to the control, but with a lower seedling height, so it can be inferred that they are more compact seedlings with more cells per cm of stem height.
As observed in the present research, there is a reduction in seedling height with paclobutrazol applications and low population densities, which is documented in the literature. Giovinazzo and Souza-Machado (2001) found that 50 ppm of paclobutrazol applied using the drench method at sowing reduced the size of tomato seedlings by up to 43 %. Seleguini, Vendruscolo, Cardoso-Campos, and de Araujo-Farias (2016) report a 10-cm decrease in tomato plant height when 50 ppm of paclobutrazol was foliarly applied at 15 das. Wien (1999) points out that, with a low population density in the seedbed, competition for light between seedlings is delayed and seedlings are less elongated. This occurred in the present experiment; furthermore, the decrease in leaf area per seedling, by reducing competition for light, stimulated the photosynthesis rate, which was reflected in a significant thickening of the stem and with it a greater seedling dry weight. Under these conditions, at the time of transplant, the seedlings are better prepared to withstand water stress and mechanical damage that usually occur at that time.
The reduction in height, leaf area and leaf area ratio, as well as the increase in dry weight and stem thickness in seedlings to prolong transplantation to 60 das without subsequent negative effects on yield and fruit quality, are considered very important from an agronomic and economic point of view. This is because it is possible to shorten the cycle, from transplanting to the end of harvest, to less than 90 days in the production system with pruning to three clusters per plant, enabling intensive greenhouse production to obtain four growing cycles per year instead of three (25 % more yield than what has been achieved so far with this production system). It should be noted that, in this experiment, the harvest of the first ripe fruits began at 112 das, and the last cut was at 145 das, that is, 85 days after transplanting. Considering the above, treatments combining three applications (20, 40 and 60 das) of paclobutrazol (25 or 50 ppm) and the lower population density (150 seedlings·m-2) produced seedlings with higher agronomic quality for transplanting at 60 das.
Yield and its components
The analysis of variance carried out on variables related to tomato fruit yield and its components (data not shown) indicated significant differences between population density treatments for number of flowers and yield per plant, and highly significant differences between paclobutrazol applications for number of flowers, number of fruits harvested and yield per plant. In no case was the density x paclobutrazol interaction significant. Coefficients of variation were very low (between 4 and 7 %), which contributed to the detection of significant differences even with close numerical values between the variables of the treatments.
Comparisons of means (Table 3) show that, in the average of the paclobutrazol application treatments, with a density of 300 seedlings·m-2, it was possible to produce one more flower per plant than with 150 seedlings·m-2, a difference that was significant; however, the number of fruits was statistically equal between the two densities. Neither did the average fruit weight show differences between densities, but the yield per plant, although with little numerical difference, was significantly higher when the seedbed was managed at the higher density (300 seedlings·m-2). No satisfactory explanation was found for this result. Possibly in the low-density treatment, because of the greater spacing between seedlings, a microclimate with a higher wind speed, higher temperature and lower relative humidity was formed within the canopy, which can negatively affect the number of flowers that reach anthesis. In any case, a more in-depth study aimed at clarifying what happened is advisable.
|Treatment||Number of flowers per plant||Number of fruits per plant||Fruit weight (g)||Yield (g·plant-1)|
|150 seedlings·m-2||21.7 bz||16.7 a||105 a||1754 b|
|300 seedlings·m-2||22.7 a||17.1 a||110 a||1864 a|
Regarding the comparison of means of the paclobutrazol application treatments in the average of the densities (Table 4), it was found that the triple application (20, 30 and 40 das) of 50 ppm of paclobutrazol caused the formation of more flowers per plant relative to the control (23.6 against 21.8 flowers), a difference that was significant. Similarly, in several of the paclobutrazol treatments, particularly those with triple applications of 25 and 50 ppm, significantly more fruits were produced per plant than in the control (at least two more fruits per plant), while the average fruit weight was similar in all treatments, including the control (Table 4). As a consequence of the greater number of fruits, the yield per plant was statistically higher than the control in several of the evaluated treatments, but the triple application of 25 ppm of paclobutrazol, which had a yield per plant of almost 300 g more than the control, stands out.
|Treatment||Number of flowers per plant||Number of fruits per plant||Fruit weight (g)||Yield (g·plant-1)|
|Control without application||21.6 bz||15.2 c||108 a||1626 b|
|50 ppm, 20 das||21.7 ab||15.8 bc||110 a||1735 ab|
|50 ppm, 30 das||22.2 ab||17.1 abc||105 a||1834 a|
|50 ppm, 40 das||22.0 ab||17.3 ab||108 a||1875 a|
|25 ppm, 20 and 40 das||22.1 ab||17.0 abc||108 a||1834 a|
|50 ppm, 20 and 40 das||22.1 ab||17.3 ab||104 a||1804 ab|
|25 ppm, 20, 30 and 40 das||22.9 ab||17.9 a||108 a||1913 a|
|50 ppm, 20, 30 and 40 das||23.4 a||17.3 ab||107 a||1852 a|
According to Contreras-Magaña, Arroyo-Pozos, Ayala-Arreola, Sánchez-del Castillo, and Moreno-Pérez (2013), and Heuvelink and Okello (2018), in the tomato flower initiation period, young leaves in the growth phase leave fewer photoassimilates available for the inflorescences that are just starting, limiting the number of flowers that can form per inflorescence. Each flower primordium needs a daily minimum of photoassimilates for its growth; if there is not that minimum number of assimilates, some or several primordia abort in favor of the others that are growing at the same time.
Dikshit, Bennett, Precheur, Kleinhenz, and Riedel (2004) applied paclobutrazol to tomato seeds and promoted a greater number of flowers in the first two inflorescences formed. In the present study, significant increases were found in the number of flowers per plant in the treatments with triple application of paclobutrazol with respect to the control, especially with 50 ppm. The greatest effect seems to be in the third inflorescence (data not shown), suggesting that probably earlier applications are required to affect the first and second inflorescence. However, with transplantation occurring up to 60 das, the cycle from transplant to end of harvest took place over 85 days, which makes it possible to obtain up to four growing cycles per year in the greenhouse.
Based on results obtained on the management of tomato cultivation with high population densities and pruning to three clusters per plant, in which the feasibility of achieving three growing cycles per year and high annual productivity have been demonstrated (Sánchez-del Castillo, Moreno-Pérez, Coatzín-Ramírez, Colinas-León, & Peña-Lomelí, 2010; Sánchez-del Castillo et al., 2012; Sánchez-del Castillo, Bastida-Cañada, Moreno-Pérez, Contreras-Magaña, & Sahagún-Castellano, 2014), a commercial validation of this system was carried out. For this, this system was compared with conventional management; this was done under a technology transfer project between tomato producers in the State of Puebla and the Universidad Autónoma Chapingo during 2014 and 2015. In the management with high population densities, the average yield per cycle was 142.13 t·ha-1 (426.4 t·ha-1·year-1, 126 t more per year than the conventional system), with an average production cost per cycle of $770,000.00 MXN ($2,310,000.00 MXN per year), which represented an average net profit of $1,041,340.00 MXN per cycle ($3,124,020.00 MXN per year, against $1,823,600.00 MXN of net profit per hectare under the conventional system).
The importance of achieving one more production cycle per year lies in the ability to generate additional net profits for producers, which are about a million pesos more per hectare than those obtained with only three cycles.
The two tested population densities enabled successful transplantation up to 60 das. With the density of 150 seedlings·m-2, seedlings with lower height, larger stem diameter, higher dry weight and lower leaf area ratio were obtained; however, at the end of the growing cycle, the number of flowers and the final yield per plant were higher in the plants that grew in the seedbed at a higher density (300 seedlings·m-2), so it is considered the most suitable for seedbed management.
The treatments involving three applications of paclobutrazol (20, 30 and 40 das) decreased seedling height, leaf area and leaf area ratio with respect to the control. In addition, the triple application produced an increase of two flowers and two more fruits per plant compared to the control, which was reflected in 287 g of additional yield per plant at the end of the cycle.
With transplantation up to 60 das, the end of the harvest occurred in 85 days, which makes it possible to obtain up to four growing cycles per year in the greenhouse, and with it 25 % more yield and annual net profits than what have been obtained so far with three cycles.