Site and plants

This study was conducted on mango cv. Tommy Atkins at a commercial orchard located at the municipality of San Francisco Ixhuatán, Oaxaca, Mexico (16°19’41.59’’N, 94°28’55.76’’O) during the 2016-2017 season. The plants were seven years old and planted at distance of 10x10m without irrigation, and prevalent disease and pest management were followed according to agronomic management recommended for the region.

Experimental design and treatments

The experimental design was a completely random. The experimental unit was integrated by two mango trees and each treatment assessed in triplicate. The treatments were the following: (1) water (control), (2) 2 % KNO₃ (PN 2 %), (3) 4 % KNO₃ (PN 4 %), (4) 6 % KNO₃ (PN 6 %), (5) paclobutrazol at 1 g of a.i. m^-1 of canopy diameter plus 2 % KNO₃ (PBZ + PN 2 %), (6) paclobutrazol at 1 g of a.i. m^-1 of canopy diameter plus 4 % KNO₃ (PBZ + PN 4 %), (7) paclobutrazol at a 1 g of a.i. m^-1 of canopy diameter plus 6 %KNO₃ (PBZ + PN 6 %), (8) 2 % NH-₄NO₃ (AN 2 %), (9) 3 % NH₄NO₃ (AN 3 %), (10) 4 % NH₄NO₃ (AN 4%), (11) paclobutrazol at 1 g of a.i. m^-1 of canopy diameter plus 2 % NH₄NO₃ (PBZ + AN 2 %), (12) paclobutrazol at 1 g of a.i. m^-1 of canopy diameter plus 3 % NH₄NO₃(PBZ + AN 3 %), (13) paclobutrazol at 1 g of a.i. m^-1 of canopy diameter plus 4 % NH₄NO₃ (PBZ + AN 4 %), and (14) paclobutrazol at 1 g of a.i. m^-1 of canopy diameter (PBZ).

Treatments preparation and application

The application of PBZ (4-chlorophenyl-4, 4-dimeth-yl-2 (1-H-1,2,4-triazole-IL) pentan-3-0, (Syngenta^®, USA) to the soil (in drencher) was carried out when the second vegetative flush emerged after the previous harvest which was on July 6, 2016. First, a representative random sampling of tree canopy diameter was carried out in order to calculate an average of canopy diameter (it was obtained a tree canopy diameter of 3.4 m). The dose of PBZ was applied once when soil was at field capacity, at a concentration of 1 g of a.i. m^-1 of canopy diameter (^{Kumbhar et al., 2009}). The aqueous solution was prepared using 3.4 g of a.i. of PBZ in 5 L of clean water, spreading the aqueous solution volume evenly into 5 ditches of 30 cm long by 10 cm depth, located at one meter far from trunk.

The potassium (KNO₃) and ammonium (NN₄NO₃) nitrate applications were carried out when the buds of the last vegetative flush reached the second stage of development (^{Pérez-Barraza et al., 2009}). Both nitrates were applied twice; the first one on September 16 and the second one on September 21, 2016. These treatments were applied with a power backpack sprayer ensuring a homogenous application over the mango tree foliage until the solutions run-off.

Assessment of floral induction variables

When the inflorescence emerged, days to flowering (d) and panicle number were recorded. Days to flowering were obtained counting the days elapse after the last application of nitrate to blooming beginning. Panicle number were registered each seven days at morning since the first panicle was observed, quantifying the total number of panicles per tree.

Production variables at harvest

Fifteen panicles were tagged on each mango tree at chest height distributed uniformly around the mango tree. The fruits from the tagged panicles were harvested 105-115 days after the flowering and transported immediately to the Food Laboratory at Colegio de Postgraduados Campus Tabasco for the assessment of the production variables. Before measuring the production variables, the fruit was washed with tap water to remove mango sap and dust, selected, randomized and then air-dried at room temperature.

Fruit weight and size

Twenty mangoes randomly selected from the harvested fruit of each treatment were used to determine fruit weight and size (equatorial and polar diameter). The fruit size (mm) and weight (g) were measured with a digital caliper CALDI-6MP^® and with a Pioneer^® electronic balance (OHAUS Corporation, U.SA), respectively.

Statistical analysis

Data were statistically analyzed by ANOVA employing SAS software (SAS^® for Windows, 9.0.) and significance was for p ≤ 0.05. When appropriated, the Fisher’s Protected Least Significant Difference test was used to separate mean values.

Effect of nitrates alone or with PBZ on the floral induction and panicle development

All treatments with potassium nitrate significantly reduced the number of days for flowering beginning (p < 0.05) with respect to the control treatment, except the PN 2 % and PBZ treatments. The potassium nitrate with PBZ (PBZ + PN 2 %, PBZ + PN 4 % and PBZ + PN 6 %) treatments were significantly the most effective in reducing the number of days for flowering beginning, with value of 16.0 d, while in the control treatment was observed a number of days to the flowering beginning of 107.0 d after the last application of treatment, followed by the PN 4 % and PN 6 % treatments, with a number of days to the flowerimg beginning of 54.5 and 61.5 d after last application of treatment (Table 1).

In contrast, ^{Yeshitela et al. (2005)} observed flowering 85 days after foliar application of KNO₃ at 4 %, but a higher panicle number was observed in mango cv. Tommy Atkins. Likewise, ^{Sarker and Rahim (2013)} reported that plants sprayed with KNO₃ at 4 % expressed earlier panicle appearance by 17 days respect to control plants in cv. Amrapali. In addition, ^{Maloba et al. (2017)} found that PN at 4 % significantly shortened flowering time on cv. ‘Ngowe’ and ‘Apple’ of mango. The PBZ application to soil followed by PN (2, 4 and 6 %) foliar spray allowed the first panicles 90 days (16 to 20 days after the last nitrate spraying) before natural flowering (non-treated trees; Table 1).

PBZ plus PN applications induced flowering, although not all dosage produced a similar number of developed panicles, in fact, the best treatment in this research work was the PBZ plus PN 6 %, since not only induce early flowering but also reached a major number of panicles (Table 2).

These results are in agreement with those obtained by ^{Oosthuyse (2015)} who reported a similar effect on plants treated with PBZ plus PN in mango cv. Nam Doc Mai Si Thong in Thailand. Likewise, paclobutrazol (soil application) at 1.0 g a.i m^-1 of canopy diameter plus PN at 2 % recorded the highest flowering intensity (^{Gopu et al., 2017}). Also these same results in this study, were similar to those revealed by ^{Rebolledo-Martinez et al. (2008)} and ^{Pérez-Barraza et al. (2009)} who found that the application of PBZ plus PN and PBZ plus PN at 4 % allowed flowering 51 days before the regular flowering time of non-treated plants in cv. Manila in tropical conditions of Nayarit Mexico and 37 days respect to the natural flowering in cv. Manila, respectively. It is known that nitrates have action on the bud dormancy breaking (^{Ionescu et al., 2017}) by increasing the activity of nitrate reductase and stimulating the production of ethylene (^{Patoliya et al., 2017}). We confirmed in this research work that the application of PBZ in sequential combination with PN suggests a synergistic effect between them, which is in agreement with ^{Rebolledo-Martínez et al. (2008)} who mentioned that the earlier appearance of the inflorescences in treated plants with PBZ plus PN might be due to the synergistic action between PBZ and nitrate to induce flowering. Also, it is generally believed that PBZ is a gibberellin inhibitor reducing the vegetative promoter level and thus stimulates flowering shoots of fruit crops (^{Guevara et al., 2012}; ^{Kishore et al., 2015}; ^{Burondkar et al., 2016}). Likewise, more recent studies revealed that mango flowering coincides with increase in nonenzymatic and enzymatic antioxidant activities, and a high antioxidant status induced by paclobutrazol is responsible for its floral responses (^{Bindu et al., 2018}). Potassium nitrate stimulates early flowering and increase number of panicles in trees growing in tropical and subtropical regions (^{Tiwari et al., 2018}). The efficacy of flowering inducing chemicals is dependent on several factors including mango variety, dose, time of application and stage of development, among others.

In this research the PBZ treatment applied alone did not shorten the period for the beginning of flowering on Tommy Atkins mango with respect to the control treatment (Table 1). Our results are in agreement with other researcher. For instance, soil application of PBZ to 3 mL m^-1 canopy diameter did not significantly affect the floral induction (138.3 days) when compared to the control treatment (152.6 days) in mango cv. Raspuri (^{Srilatha et al., 2015}). Moreover, several treatments with PBZ applied in soil drencher did not induce flowering in mango cv. Dashehari (^{Narvariya et al., 2014}). Conversely, treatments of PBZ caused earlier panicle emergence in mango cv. Rosa and Alphonso compared with control treatment (^{Shinde et al., 2015}).

Regarding the effect of ammonium nitrate treatments to induce flowering, the PBZ + AN 3 %, PBZ + AN 2 %, and PBZ + AN 4 % treatments significantly induced flowering 17.8, 18.3, and 19.5 days to flowering after last treatment aplication, respectively, compared to the control treatment (107.0 days to flowering). Contrarily, ammonium nitrate treatments applied alone had no effect (p < 0.05) on the floral induction in mango cv. Tommy Atkins (Table 1). These findings coincide with those reported by ^{Salazar-García et al. (2000)} in cv. Tommy Atkins. On the contrary, plants treated with PBZ plus AN (2, 3 or 4 %) emitted early panicles similar to those observed with PBZ plus PN, but the number of emerged inflorescences was not different to non-treated plants. This latter differs with the reported by ^{Silva et al. (2013)} who informed that plants treated with PBZ plus AN increased 85 % the panicles appearance when compared to the non-treated plants.

For the number of developed panicles, in our study, there were highly significant differences (p < 0.05) for treatments with PN. Only the PBZ + PN 6 % treatment significantly increased the number of developed panicles (288.5) with respect to the control treatment. Although the number of developed panicles in PBZ + PN 2 % and PBZ + 4 % treatments were higher than those of the control treatment, no significant difference was observed among them, with values of 156.6, 103.0 and 136.0 developed panicles, respectively. A minor number of panicles was observed with the PN 2 %, PN 4 %, and PN 6 % treatments (Table 2). On the other hand, the PBZ + AN 4 % treatment produced the higher number of developed panicles (136.5), however it was statistically equal to that obtained by the control treatment (136.0). Trees treated with the different concentrations of AN (AN 2 %, AN 3 % and AN 4 %) without application of PBZ emitted a sparse flowering (Table 2).

Effect of nitrates alone or with paclobutrazol on produc-tion variables at harvest

Treatments with potassium nitrate had a significant effect (p < 0.05) on weight and size of mango fruit (production parameters) cv. Tommy Atkins at harvest. All treatments with potassium nitrate significantly reduced the equatorial diameter of mango fruit with respect to the control treatment (equatorial diameter of 160.65 mm), with the exception of the PN 2 % treatment (equatorial diameter of 171.34 mm). The lowest equatorial diameters of mango fruit were observed with the PBZ + PN 2 %, PBZ + 4 %, and PBZ + 6 % treatments, with values of 91.94, 92.99 and 90.61 mm of equatorial diameter (Figure 1). Moreover, the effect of treatments with potassium nitrate had a similar trend on the polar diameter of mango fruit with respect to the control treatment. The highest polar diameter of mango fruit was observed with the PN 2 % treatment, with a value of 183.51 mm. Conversely, the lowest polar diameter was observed in PBZ + PN 2 %, PBZ + PN 4 %, and PBZ + PN 6 % treatments, with values of 114.20, 108.64 and 111.11 mm, respectively. Surprising, the highest fruit weight (665.9 g) was observed with the control treatment, while the rest of treatments with potassium nitrate reduced or not the fruit weight of mango. The lowest fruit weight was registered with the PBZ + PN 2 %, PBZ + PN 4 %, and PBZ + PN 6 % treatments, with values of 502.89, 469.54 and 471.12 g, respectively (Figure 1C). Even though the fruit weight in the PN 2 % treatment was statistically equal to the control treatment, the equatorial and polar diameters (171 and 183 mm, respectively) were significantly larger compared to the control treatment (160 and 169 mm, respectively) (Figure 1A-B).

Figure 1 Effect of potassium nitrate alone or with paclobutrazol on the fruit weight and size of mango cv. Tommy Atkins: A, equatorial diameter (mm); B, polar diameter (mm) and C, fruit weight (g). Bars with different letters are statistically different (Fisher, p ≤ 0.05) (n = 20).
Figura 1. Efecto de nitrato de potasio sólo o con paclobutrazol en el peso y tamaño de frutos de mango cv. Tommy Atkins: A, diámetro ecuatorial (mm); B, diámetro polar (mm) y C, peso de fruto (g). Barras con diferentes letras son estadísticamente diferentes (Fisher, p ≤ 0.05) (n = 20).  

Treatments with ammonium nitrate had a significant effect (p < 0.05) on the weight and size of mango fruit (production variables) cv. Tommy Atkins at harvest. All ammonium nitrate treatments significantly reduced the equatorial diameter of mango fruit with respect to the control treatment (equatorial diameter of 160.65 mm), with the exception of the PBZ + AN 3 % treatment (equatorial diameter of 156.21 mm). The lowest equatorial diameters of mango fruit were observed with the PBZ + AN 2 % and PBZ + AN 4 % treatments, with values of 101.1 and 102.0 mm of equatorial diameter (Figure 2A). Moreover, the effect of treatments with ammonium nitrate had a similar trend on the polar diameter of mango fruit with respect to the control treatment. The lowest polar diameters of mango fruit were observed with the PBZ + AN 2 % and PBZ + AN 4 % treatments, with values of 113.5 and 119.4 mm, respectively (Figure 2B). Likewise, the lowest fruit weight was observed in the PBZ + AN 4 %, AN 4 %, and PBZ + AN 2 % treatments, with values of 518.34, 567.88 and 583.54 g, respectively (Figure 2C), while, the mango fruit weight in the rest of the treatments with ammonium nitrate did not significantly differ to the control treatment (665.86 g) (Figure 2C). According to our above mentioned results, the sequential treatments of PBZ plus PN or AN significantly reduced the mango cv. Tommy Atkins fruit weight and size (Figure 1 and 2). These results agree with those found by ^{Oosthuyse and Jacobs (1997)} in cv. Tommy Atkins who conclude that fruit weight decrease as PBZ concentration increase, they interpret this fact as due to fruit overcrowding resulting in more competition for the available resources. These effects are also reported in other horticultural crops as cucumber, tomatoes and avocado (^{Magnitskiy et al., 2006}). Although in mango cv. Manila, Sensation and Kent the weight of fruit was not affected by PBZ treatments (^{Rebolledo-Martínez et al., 2008}). Likewise, we found that PN 2 % or PN 4 % did not affect the fruit weight (Figure 1) which coincide with those reported by ^{Ataide and Jose (2000)} and ^{Burondkar et al. (2013)} where foliar spray of PN at 3 % did not influence the fruit weight on ‘Tommy Atkins’ and ‘Alphonso’ mango, respectively.

Figure 2 Effect of ammonium nitrate alone or with paclobutrazol on the weight and size of mango fruit cv. Tommy Atkins: A, equatorial diameter (mm); B, Polar diameter (mm) and C, weight of fruits (g). Bars with different letter are statistically different (Fisher, p ˂ 0.05) (n = 20).
Figura 2. Efecto de nitrato de amonio solo o con paclobutrazol en el peso y tamaño de fruto de mango cv. Tommy Atkins: A, diámetro ecuatorial (mm); B, diámetro polar (mm) y C, peso de fruto (g). Barras con diferentes letras son estadísticamente diferentes (Fisher, p ˂ 0.05) (n = 20). 

Another fact in this research is that the bigger weight and size of fruit from the non-induced plant could be attributed to the water supply by rainfall and solar radiation on the period of fruit development on the trees concurring with the rainy season from the region (^{Spreer et al., 2009}). followed by reviews of plant water relations, water requirements, water productivity and water management. This long-lived tree is well adapted to a wide range of tropical and subtropical environments. In the low-latitude tropics, flowering is initiated after a period of water stress (at least six weeks durationEven though the fruit weight was affected negatively by several flowering inducer treatments, fruit were in the acceptance category of the required by the Mexican norms of mango fruit quality.