Production

In 2010, without fertilization, there were significant differences of production between localities. In the Las Palmas orchard there were differences in production, where the trees selected to apply high and normal doses had a higher yield compared to the control, but in general had an average of 73.8 kg tree^-1. The following years, already applied fertilization treatments, a trend of increase in productivity was observed in all the treatments, where the control always presented smaller production.

In 2011 production increased and in 2012 decreased, but recovered in 2013. ^{Salazar et al. (2003)} applied the FSE in avocado cultivation, obtaining the initial year an intense increase in yield, but the subsequent years showed an effect of moderate alternation in yield. On the other hand, in the orchard of Buenavista there were no differences in yield for 2010, with a mean of 131.8 kg tree^-1, but in the following years the normal dose had higher production with respect to the high dose and the control, the latter with the lowest production (Table 1). This indicates that the normal dose is sufficient to improve productivity, since the high dose may not be beneficial and represent an excessive expenditure of fertilizers, but both doses are better than the control. Also, ^{Salazar et al. (2014)} in ‘Kent’ mango obtained the highest fruit yield (138.2 kg tree^-1) with a normal dose, whereas for ‘Tommy A’ mango the highest yield was obtained with the high dose.

Fruit size in length, diameter, weight and caliber

The fruits of 2010 without application of treatments had an average weight of 525.8 g and 77% of the fruits had calibres between 8 and 10. In Buenavista fruits were smaller than those of Las Palmas, with 478.6 g and 573 g respectively. Table 2 shows that the effect of year, locality and dose factors had significant differences in size variables except caliber. The gauge is considered to be insignificant because of the wide weight range corresponding to each gauge. By locality, the fruits of Buenavista were superior (427.3 g) with respect to those of Las Palmas (341 g), indicating a positive effect in increase of size, with respect to the base year.

In the years of fertilization, there is a tendency to increase in size, where 2014 showed the fruits with higher values in length, diameter and weight, this can indicate a favorable effect of fertilization year after year on the size of the fruits. In the fertilization treatments in terms of length and caliber, there were no significant differences, but there were differences in diameter and weight, where the control had the lowest values compared to the normal and high doses, which were statistically no differences, but numerically the normal dose had fruits of greater weight and consequently smaller caliber, this indicates that the high dose has a less favorable effect and that a high dose is not necessary to obtain fruits of good size.

This agrees with ^{Salazar et al. (2014)} where in mango ‘Kent’ with normal and high dose obtained greater production of fruits caliber 12 (350-414 g). According to the Mexican quality standard for fresh export mango the acceptable marketing gauges are between 7 and 14 (648 and 324 g); thus, 100% of the fruits with fertilization had acceptable caliber’s, within a median size.

Weight loss

The weight loss (PP) in fruits from 2010 to 13 DDC did not have significant differences between localities, with 8.2% for Las Palmas and 7.6% in Buenavista. The following years with the application of fertilization treatments, the to consumption was significant for the factors year, locality and treatments (Table 3). By locality, the fruits of Buenavista had greater PP. From harvest to maturity of consumption (12 days), PP in fruits had a tendency to increase, with significant differences between years, where the fruits of 2014 had lower PP (5.2%) and those of 2011 a higher PP (7.4%) to consumption with 12 DDC. At 7 DDC in 2011, a PP of 5.4% was observed, while the fruits of 2012 and 2014 reached those percentages of PP at 12 DDC, indicating an increase in the shelf life of the fruits.

In mango ‘Kent’ for 12 days at 20 °C a 2.5% PP (^{Siller et al., 2009}) was reported, much lower than that found in this study, which is attributed to storage temperature. In ‘Haden’ mango, 5.8% PP was reported at 9 days at room temperature (^{Romero et al., 2006}). The PP may be due mainly to perspiration occurring through lenticels, as well as increased respiratory metabolism and ethylene production caused by stressful treatment (^{Mitra and Baldwin, 1997}).

Due to the fertilization dose, the control had less weight loss than the normal and high dose, these without significant differences. The above could indicate that the normal and high dose increases the metabolic activity during the maturation process, contrary to what reported by ^{García et al. (2015)} in ‘Kent’ mango with high fertilization and control, respiratory intensity was lower. Other studies also report lower respiration during maturation and weight loss by application of Ca in “Haden” mango (^{Mulkay et al., 2003}; ^{Romero et al., 2006}).

On the other hand, fruits that present a loss of weight of more than 5%, can produce a commercially unacceptable product (^{Luna et al., 2006}), a value that was exceeded in this study; however, external appearance must be considered to determine acceptability. Although the external appearance data were not shown, it remained excellent to good (0 - 1) from harvest to 12 DDC in consumption and in the treatments there were no significant differences in appearance, being good to excellent from harvest until consumption.

Color of shell

Mango color is one of the most important physicochemical characteristics in the ripening process and one of the main criteria of consumer acceptance (^{Slaughter, 2009}). In the fruits of 2010 the development of color in shell until maturity of consumption (a* = 6.9) was low in comparison with the following years where the fertilization treatments were applied. The trend in the development of color in the shell was to increase the value of a*, starting at negative values and​ reaching positive values for the degradation of chlorophyll and appearance of the carotenoids that provide the yellow colors (^{Ornelas et al ., 2008}). In the Table 4 shows that initially and at 7 DDC there were no significant differences by locality, but at maturity of consumption (12 DDC) there was greater degradation of the green color in the fruits of Buenavista. In the year factor if there were differences in color in the crop, but at maturity of consumption reached similar values.

As for the effect of the doses, on consumption (12 DDC) the high dose presented greater intensity in the green color than the other treatments, which can be attributed to the high amount of nutrients. It has been reported in studies that the presence of nitrogen at the levels of normal and high fertilization in ‘Kent’ mango increased the green tonality of the background color of the fruits (^{García et al., 2015}) and that mango fertilization ‘Tommy Atkins’ with calcium sulfate contributed to maintain the green color of the shell (^{Arizaleta et al., 2001}).

Pulp color

The color change in pulp is due to the appearance of new carotenoid compounds that cause the color to change constantly during the maturation process giving a yellow orange pigmentation of the mesocarp of the mango (^{Ornelas et al., 2008}). Thus, in 2010 it began with an average of H= 83.25°, with no differences between localities, reaching H= 63.7° at maturity of consumption. The decrease in Hue values from harvest to consumption indicates the increase in the intensity of the yellow-orange color in pulp. For the years 2011, 2012 and 2014 with the application of the doses, to harvest the fruits reached the optimum degree, with yellow egg color in 100% of the pulp (H> 83°). Between years of fertilization there were significant differences, with a tendency to decrease color intensity in 2014, presenting higher values, which would indicate that the consecutive fertilization delayed the development of pigmentation in pulp (Cuadro 5).

The fruits of Las Palmas presented greater development of yellow orange color than those of Buenavista to maturity of consumption (12 DDC). Regarding the doses there were significant differences, the normal developed at all stages of sampling less intensity in pulp color than the discharge and control; but the numerical difference is only two degrees Hue, which visually is imperceptible (Table 5). Larger values (79 ± 3.3°) were reported on ‘Kent’ mango at 12 DDC at 20 °C (^{Siller et al., 2009}). In what reported by ^{García et al. (2015)} in mango ‘Kent’ there were no differences between the three levels of fertilization, indicating that there was no effect of fertilization.

Firmness of pulp

In the process of maturation of the mango pulp until maturity of consumption (12 DDC), there was a tendency to decrease the firmness, which is attributed to the degradation of the amyloplasts by hydrolysis of the starches produced in photosynthesis, which when hydrolyzed produce low-molecular-weight carbohydrates soluble in water, which is reflected in the increase of soluble solids (^{Yashoda et al., 2007}). The firmness of the fruits of 2010 did not have significant differences between localities for both harvest and consumption, with 199.8 N and 17.15 N respectively.

In the years with the application of the fertilization doses (Table 6), there were significant differences by locality, where the fruits of Buenavista had more firmness than those of Las Palmas. In the year factor there were significant differences, in 2014 the fruits always maintained more firmness with respect to 2011 and 2012 in all the samplings. This would indicate that as in shell color, there is a cumulative residual effect of nutrients with the application of fertilization with the years that could delay the softening process of the fruits. However, at fertilization doses there were significant differences at harvest, the normal dose had greater firmness than the high dose and control, but all were above the minimum acceptable value of 121.6 N (^{Báez, 1998}).

A maturity of control as well as the high dose reached values of firmness below acceptable quality (17.15 - 19.6

N) according to ^{Mitcham and MacDonald (1992)}, while the normal dose maintained the fruits with greater firmness, which would indicate a favorable effect on the shelf life of the fruits. These results were very different from those reported by ^{García et al. (2015)} in ‘Kent’ mango, where the level of high fertilization and control delayed firmness significantly above the normal level in the first five days, but at eight days the control firmness was greater.

Total soluble solids

The total soluble solids (SST) is one of the most interesting physicochemical parameters in the ripening process, as the fruit becomes sweeter and therefore more palatable to the consumer (^{Quintero et al., 2013}). In the year 2010 the TSS from harvest to consumption fluctuated from 7.7 to 17.1 °Brix, while in the following years with the application of the fertilization doses the minimum value at harvest of 7.4 °Brix was exceeded (^{Báez, 1998}), having values between 8.6 and 9.6 °Brix, with significant differences by locality and fertilization treatments, but not for the evaluated years.

The trend of SST content from harvest to maturity always showed a gradual increase (Table 7). The fruits of the locality of Las Palmas from harvest (0 DDC) to maturity of consumption (12 DDC) had higher SST content than Buenavista. At the fertilization doses, the control had the fruits with the highest SST content, reaching 18.2 °Brix. The opposite was for the normal and high dose that developed the fruits with lower SST content, 17.3 and 17.4 °Brix respectively, which were higher than the 13.7 °Brix reported by ^{Siller et al. (2009)} at 12 DDC at 20 °C. On the other hand, ^{García et al. (2015)} in another similar study obtained results contrary to those obtained in this study, since normal-level fertilization had a greater accumulation of SST than the control.

In the years of application of the fertilization doses, there were no significant differences both initially and in consumption. These results show that there is no effect per fertilization dose that is reflected in the development of SST, but the opposite, is diminished for both the normal dose and the high dose.