Selected orchards

A commercial orchard with fertigation for each producing region of the states of Jalisco (Jal), Michoacan (Mich) and Nayarit (Nay), which presented different altitudinal and climate characteristics, was worked. The region of “hot” climate is the municipality of Tepic, Nayarit, with the orchard “El Parejo”, warm subhumid climate, altitude of 1 151 m, average temperature of 21.1 °C and annual rainfall of 1 287 mm. In the “intermediate” region, the orchard “Paso of Carretas” was selected, in the municipality of Gómez Farias, Jalisco, with a subhumid climate, altitude of 2 180 m, average temperature of 19.8 °C and annual rainfall of 717.8 mm. The region with “cold” climate is Uruapan, Michoacán with the orchard “El Parejo”, in temperate climate, altitude of 1 579 m, average temperature of 19.3 °C and annual rainfall of 1 427.1 mm.

Fruit sampling

In each orchard a group of 30 trees (5 replicates of 6 trees each) were selected, in which fruits corresponding to the main flowering occurred in january-february 2015. During 2015, fruits were sampled in several stages of development, according to its diameter (ø): olive stage (20-30 mm ø; Nay: 15 Apr, Jal: 15 May, Mich: 29 Mar), stage I (35-45 mm ø; Nay: 22 May, Jal: 03 Jun , Mich., 25 May), stage II (50-60 ø; Nay: 16 Jun, Jal: 15 Jul, Mich: 22 Jun), stage III (60-70 mm ø; Nay: 06 Aug, Jal: 25 Aug, Mich., 29 Jul), and harvest, when the dry matter (MS) content of the pulp (mesocarp) reached ≥ 21.5% (Nay: 10 Sep, Jal: 12 Nov, Mich: 29 Sep).

At each sampling, the fruits were washed with distilled tap water and then weighed and measured.

From each tree, at least four fruits were collected per sampling date and to reduce variations due to sun exposure, these were selected to the southwest of the middle and outer part of the tree crown. The skin was separated with a vegetable peeler and stored at -80 °C until analyzed. For the determination of phytochemicals composites composed by the skin of the fruits collected in the six trees that conformed each repetition were integrated.

Phytochemical analysis

For the quantification of phytochemicals the total phenolics (CFT), total chlorophylls (CT) and total carotenoids (CAT) of the skin were extracted with an acetone: water solution (80:20) (^{Rodriguez et al., 2011}). The CFT were quantified using the Folin-Ciocalteau method (^{Stintzing et al., 2005}) and expressed in mg equivalents of gallic acid per gram of dry skin (mg EAG g^-1 PS). The chlorophylls were determined according to ^{Donetti and Terry (2012)} and expressed in mg g^-1 PS. The concentration of CAT (mg g^-1 PS) was calculated according to ^{Lichtenthaler and Buschmann (2001)}. The lignin concentration was determined on dry and ground skin subjected to acid hydrolysis with 72% sulfuric acid (^{Ankom Technology, 2013}) and expressed as a percentage.

Thickness of skin

Five fruits were collected per repetition and portions of skin were taken from the upper (near the pedicel), middle and lower parts of the fruit. The skin samples were fixed in FAA (5 formaldehyde: 5 glacial acetic acid: 90 ethanol, v/v) and then dehydrated by sequential transfer to sets of ethanol solution (70%, 85% and 95%), infiltrated and polymerized in glycol methacrylate (Leica embedding Kit Historesin). The polymerized skin samples were cross-sectioned to 5-8 micrometers in thickness on a rotary microtome (HM 350 S MICROM), mounted in water on slides, dried on a hot plate, dip-stained in 0.05% toluidine blue and were photographed with a composite microscope. The obtained sections of skin were measured with a micrometer, installed in the eyepiece of the microscope, the thickness of the skin and the cork, in the flat areas and where there were protuberances, through the development of the fruit.

Nutritional analysis and dry matter

The skin of five fruits was used per replicate. The nutrient quantification was done in dehydrated skin in an oven with forced air at 65 °C and pulverized in a 40 mesh stainless steel mill. The concentrations of N, P, K, Ca, Mg, S, Fe, Cu, Mn, Zn and B were determined in a commercial laboratory under the NAPT (Soil Science Society of America) program. The dry matter was determined by weight difference between the fresh and dry sample and expressed as a percentage.

Measurement of roughness

Twenty fruits were selected per replicate for each stage of fruit development. Three images of each fruit were acquired: ventral lateral equatorial section, upper dorsal front and ventral frontal superior (Figure 1). A 2x2 cm (3 648 x 2 736 pixels) image was obtained on a stereoscopic microscope (Zeiss model Stemi 2000-C; Barrington, NJ, USA) with a digital still camera (Canon Power Shot G11 model; NY, USA) for each section of the fruit. Each image was incorporated into the ArcView version 3.2 geographic information system and the “Spatial Analyst” module quantified the surface with a slope greater than 20%. The roughness was expressed as the rough surface (cm²) with respect to the total area of the analyzed image (rough surface/ total area, cm² cm^-2)

Figure 1 Sections evaluated for ‘Avocado “Hass” fruit. 

Statistic analysis

The experimental design was completely random, with five replicates (6 trees per replicate). The variables included in the statistical analysis were skin dry matter (MS skin), concentration of total phenolic compounds (CFT), total chlorophylls (TC), total carotenoids (CAT), lignin, concentrations of N, P, K, Ca , Mg, S, Fe, Cu, Mn, Zn and B; thickness of the protrusions of the skin (GPRP), thickness of the flat section of the skin (GPLP), thickness of the cork protrusions (GPRCO), thickness of the flat section of the cork GPLCO), roughness of the ventral lateral equatorial section (ELV), dorsal frontal superior surface roughness (SFD) and ventral frontal superior section (SFV) roughness. A multivariate analysis by main components (ACP) was performed for the data obtained in the sampling to harvest and a global one.

With this analysis, the concentrations of phytochemicals, nuances and skin roughness of ‘Hass’ fruits in physiological maturity (MS≥ 21.5%) and fruit development were related in the three producing regions with different climate. The first two main components were selected which accumulated at least 50% of the total variation, and with characteristic values >|0.25 |. Variables were also analyzed independently by analysis of variance and comparison of means with the Waller-Duncan test (p≤ 0.05) per sampling stage and overall. To establish the possible association between the concentration of phytochemicals and nutrients with skin roughness, a correlation analysis was performed between the variables involved for each characteristic. The analysis was done with the statistical program SAS version 9.0.

Climatic characteristics

The meteorological parameters showed a marked variation, both temporal and intensity. The average annual temperature between regions differed by almost 2 °C, Nayarit had the highest value and Michoacán the lowest. The average monthly temperature was higher in Nayarit throughout the year. In Michoacán the summer was the coldest, compared to the other two regions; however, the coldest months were in Jalisco (December and January). January was the coldest month in the three regions, with variations of 1.5 °C between them.

Regarding precipitation, in the three regions it was identified that in the summer (June to September) it rains above 75% of the annual rainfall. The rainiest months were July in Nayarit and August for Jalisco and Michoacán. In the three regions, the month with the lowest rainfall was april (mean less than 5 mm). The accumulated winter rain (December, January and February) varied between 40 and 88 mm between regions. In Jalisco, the lowest annual rainfall occurred and these differences were more evident from July to September. Rainfall was the parameter with greater contrasts. The region of Michoacán received 50% more rain than Jalisco, even though it presented almost the same amount of rainy days as Nayarit. The 1 427 mm of precipitation a year in Michoacán locates it as the rainiest region (Figure 2).

Figure 2 Precipitation and monthly average temperature of the three producing regions during the development of the fruit. 

Differences in phytochemical concentration and skin roughness between producing regions and stages of fruit development

For the various parameters evaluated the ‘Hass’ skin showed differences between the producing regions and for each stage of development of the fruit. In the olive stage, the skin of Nayarit fruits had a higher concentration of total carotenoids and a lower concentration of CFT and chlorophylls. The fruits of Michoacán had the lowest concentration of lignin in the skin. The fruits of Nayarit had the lowest concentration of N, K, Ca, Mg, Cu and B; while those from Michoacan showed the highest concentration in N, Ca, Mg and Zn. The physical characteristics (thickness of the flat sections and with protuberances of the skin and the cork) did not show difference between regions. The roughness in the three zones of the analyzed fruit was smaller in the ones coming from Michoacan (Table 1).

The phase I fruits from Jalisco presented the highest values for CFT and total chlorophylls. The CAT showed the highest concentration in the skin of Nayarit fruits. Lignin showed no difference between the producing regions. The fruits of Michoacán presented a lower concentration of N, P and S, and the higher of Cu. The fruits of Nayarit showed the lowest concentration in K, Mg and Cu, while the B in Jalisco was lower than in the other producing regions. The thickness of the skin showed no difference between the regions. Regarding the roughness, ELV was higher in Nayarit (Table 1).

In stage II, Jalisco fruits differed in a higher concentration of CFT. Pigments showed no difference between regions.

The skin of the fruits of Michoacán showed the lowest concentration of P and S, and the highest of K and Mn. Nayarit had the highest concentration of Zn and B, and the lowest concentration of K and Ca. Cu showed the lowest concentration in Jalisco. The thickness of the skin only differed in the section of the protuberances in the skin, the fruits of Nayarit presented the greatest value in this characteristic. The ELV roughness was the only section of the fruit with a difference between the regions, Nayarit showed the highest level and Jalisco the lowest, Michoacán did not differ with the other two production zones (Table 1).

The skin of Jalisco fruits in Stage III differed in a higher concentration of CFT and lower in CAT, the latter without difference with Michoacán. Total chlorophylls and lignin showed no difference between regions. The fruits of Michoacan had the lowest concentration of P and Fe and the highest concentration of S and Cu, while in Jalisco, K, Ca, Mg and Mn had the highest values. The fruits of Nayarit showed the highest concentration of Fe Zn and B and the lowest of Ca. The physical characteristics differentiated the fruits of Jalisco, in which the skin showed higher GPLCO (Table 2).

At harvest, the skin of the Nayarit fruits showed a higher concentration of CFT and chlorophylls, while in Jalisco it was found in CAT and lignin. The concentration of N, P, Mg and S presented the highest values in the fruits of Jalisco, while the Fe was in the Nayarit, in this last producing area the lowest concentrations of Mg and Cu were present. The fruits of Michoacán presented the highest concentration of Cu and Ca and the lowest of P (Table 2).

In the overall analysis, Jalisco fruits showed the highest concentration of total chlorophylls. Lignin concentration showed no significant difference between producer regions Nayarit fruits had the highest concentration of CAT and B and the lowest concentration in K, Ca, Mg and Cu. The skin of Michoacán showed the highest Cu content. In Jalisco, the skin showed the highest concentration of S and the lowest concentration of B, and the flat section of cork was significantly thicker. The ELV and SFD roughness was higher in fruits from Nayarit, while those from Michoacan showed the lowest roughness in the SFD and SFV sections (Table 3).

Principal component analysis

At harvest, the first two CPs explain 65.6% of the skin variability of ‘Hass’ fruits (Table 4). The CP 1 (48.2%) included P, Mg, Zn, Fe, Ca and CFT and CT; the fruits of Nayarit differed from those from the other two regions in presenting a higher concentration of P, Zn, Fe and phytochemicals and lower Mg and Ca, whereas CP 2 included skin thicknesses, and concentrations of N and K (Figure 3).

Figure 3 Principal components for the skin of ‘Hass’ fruits to harvest from three producing regions. The letters N, J, M and the number indicate the producing region and the repetition. 

In the overall analysis the first two main components explained 53.6% of the variation among fruits (Table 4). The CP 1 (P, N, Ca, Mg, CFT and lignin) distinguished the fruits of the first two stages of development, in which fruit skin presented the highest concentrations of these nutrients and chemical compounds (Figure 4, solid line). The CP 2 included TC, MS and GPLP and differentiated fruits from Jalisco, which presented lower MS and higher CT concentration and thickness of the flat section of the skin (Figure 4, dotted line).

Figure 4 Principal components for the skin during the development of ‘Hass’ fruits from three producing regions. The letters N, J, M indicate the producing region and the numbers indicate the sampling stage and the repetition, respectively. 

Correlation analysis

Linear correlation analysis showed that CFT correlated negatively with fruit skin thickness measurements, while total chlorophylls were related only to SFV roughness. Lignin, on the other hand, showed a positive correlation with skin and cork thickness and SFD roughness. The correlation coefficients greater than |0.5| were presented between lignin, N, P, K and Ca, with thicknesses of the section with protrusions of the skin and cork (GPRP and GPRCO).

The roughness, although it was significantly related to some nutrients and phytochemicals, did it in a lower form (r < |0.44|) (Table 5).