Effect of nets on the crop environment

PAR, ambient temperature and relative humidity were different in each cover with respect to the field. Only the field and under-greenhouse measurements exceeded 1 000 μmol·m^-2·s^-1 (Figure 1), and there were reductions of between 43 and 50 % PAR with the different covers compared to that presented in the field (Table 3). The maximum PAR indices consistently occurred between 13:00 and 15:00 h, while in the colored nets these indices were between 500 and 700 μmol·m^-2·s^-1, with the red netting recording the highest indices and the blue the lowest. PAR above 1 000 μmol·m^-2·s^-1 has a direct effect on the photosynthesis rate, which negatively affects productivity, growth and yield (^{Kitta et al., 2014}). The results of the decline in PAR were notable because the shade netting had a buffering effect on the changes in this parameter over the course of the day (Figure 1).

Figure 1 Behavior of photosynthetically active radiation (PAR) under the different covers compared to the open field.  

According to ^{Retamales, Montecino, Lobos, and Rojas (2008)}, in highbush blueberry cultivation it has been observed that PAR, transmitted by white and red nets with 35 % shading, and grey netting with 50 % shading, was reduced by up to 29 %, while red netting with 50 % shading and black netting with 35 % shading decreased PAR by 41 and 47 %, respectively, compared to the field. Therefore, the use of shading nets is relevant for reducing the impact of high light intensity in open field conditions and transforming direct radiation into dispersed radiation. This allows light to penetrate the inner canopy of the plant, which prevents burns and gives a moderate cooling effect (^{Ilic et al., 2017}).

^{Shahak (2014)} points out that with any cover placed on plants, in addition to the effects on light intensity and quality, the climatic elements (solar radiation, wind speed, temperature and relative humidity) can be modified (^{Arthurs, Stamps, & Giglia, 2013}), altering the metabolism and water consumption of plants, which has a positive impact on fruit yield and quality (^{Tanny, 2012}). Temperature and relative humidity play an important role in foliar stomatal conductance and, therefore, in the rate of transpiration and photosynthesis of the plant (^{Righi, Buriol, Angelocci, Heldwein, & Tazzo, 2012}).

Effect of the nets on the crop

Highly significant (P ≤ 0.01) differences were found among covers in all the growth variables studied (Table 4). Plants grown under white netting showed the greatest statistical difference in seven of the nine variables studied (PH, BSD, AFY, NFP, FD, NS and NL), and plants grown under blue netting had the best values ​​in four of the variables (DF, FD, NS and IL); therefore, the latter can be considered as the second-best cover for piquin pepper cultivation (Table 5).

On the other hand, plants grown in the open field showed the worst results in eight of the nine variables studied (except BSD) (Table 5). Likewise, plants grown under black netting showed the lowest values in four of the analyzed variables (BSD, AFY, NFP and FD), which supports the need to look for the best conditions for the development of wild piquin pepper, in this case through the use of nets that provide the optimal ranges in the microclimatic variables.

The white netting produced a higher NFP and AFY (374.61 and 1 290.90 %, respectively), compared to open field production (Table 5). Likewise, fruits grown under white and blue nets produced larger FD (236.18 and 266.88 %, respectively) and NS (314.50 and 294.33 %, respectively), both with respect to open field production.

The decreased PAR availability may induce morphogenetic responses such as increased leaf area and lengthening of the stem and internodes, because cells expand more with low solar radiation intensities to capture light and carry out photosynthesis. This is associated with the phenomenon known as "shade avoidance", since it is influenced by the phytochromatic activity that regulates stem lengthening, by transforming non-photosynthetic plastids (etioplasts) into fully developed chloroplasts with photosynthetic function, and the leaf area, with the consequent effects on photoassimilation distribution and fruit growth (^{Bastías, Manfrini, & Grappadelli, 2012}).

In this study, the nets and greenhouse caused an increase in the final height of the piquin pepper plants (Table 5), probably in response to the reduced light (^{Salisbury & Ross, 2000}). It has been reported that when using nets the PH changes depending on the color of the net; in bell pepper the highest PH was obtained with blue nets (^{Ayala-Tafoya et al., 2015}), and in basil with black ones (^{Martínez-Gutiérrez, Nicolás-Santana, Ortiz-Hernández, Morales, & Gutiérrez-Hernández, 2016}).

Radiation is one of the factors capable of producing greater photosynthetic activity in plants; in addition, the quality and distribution of the light spectrum can influence the length of the internode and the development of flowering (^{Runkle & Heins, 2006}). In this case, this effect stood out in the white and blue nets, where an increase in PH of 114.05 and 91.53 %, respectively, was recorded compared to the open field. Additionally, the white netting increased BSD by 34.1% in contrast to the black netting, where the lowest result was obtained. This coincides with the findings of ^{Ayala-Tafoya et al. (2011)}, who report that colored shading nets increased BSD in tomato.

Principal component analysis (PCA) revealed a grouping of plants in the second quadrant (Figure 2), which indicates that they showed lower PH, NFP and FD, and higher DF (CP1), as well as greater BSD and fruit weight and lower NS (PC2), this considering the relevance of these variables for each of the components (Table 6). However, in general, the materials behaved in a dispersed manner in the analyzed conditions, which reveals that the ecotype played a determining role in the variables studied. It is important to note that a larger BSD allows the plant to have a better-developed vascular system, through which it conducts water and nutrients, which improves physiological processes (^{Bahena-Delgado, Bustos-Rangel, Broa-Rojas, & Jaime-Hernández, 2012}).

Figure 2 Graph of the principal components of the analyzed variables and types of cover. PH = plant height; BSD = basal stem diameter; DF = days to flowering; AFY = average fruit yield; NFP = number of fruits per plant; FD = fruit diameter; NS = number of seeds; IL: internode length; NL = number of leaves; WN = white netting; BLUN = blue netting; GRE = greenhouse; RN = red netting; BLAN = black netting; OF = open field. 

^{Díaz-Pérez (2014)} found an increase in the number and size of bell pepper fruit under shade conditions. On the other hand, ^{Shahak et al. (2008)} when evaluating "raschel" colored nets (red, yellow and pearl, with 30 to 40 % shading) in bell pepper obtained 115 to 135 % higher AFY compared to results obtained with black netting and in the open field. In contrast, ^{Ayala-Tafoya et al. (2015)} found no significant differences in the NFP in bell pepper when using different colored nets and production without nets.

Effect of ecotypes

All growth variables showed highly significant (P ≤ 0.01) differences among the different ecotypes analyzed (Table 4). The RTZ ecotype had the best performance in eight of the nine variables (PH, BSD, DF, AFY, NFP, FD, NS and IL), while the SAC ecotype had the worst performance in almost all variables. The community of origin of the latter is at an elevation of 365 masl, which is much lower than that of the place where the experiment was established (1 580 masl). On the other hand, the place of origin of the RTZ ecotype is at an elevation of 2 060 m, which could have influenced the adaptation to the conditions of the experiment. However, this does not coincide with what was reported by ^{Martínez-Sánchez, Pérez-Grajales, Rodríguez-Pérez, and Moreno-Pérez (2010)}, who collected wild species of Capsicum annuum at elevations of 940 to 1 600 m, and evaluated them at 2 247 m without finding significant statistical differences among species in the variables PH, BSD, NFP and fruit size.

PCA did not reveal distinct ecotype groupings, which supports the idea that the environment also plays an important role in the dispersion of data. However, some trends were observed; for example, the RTZ ecotype was grouped to a greater extent in quadrant 1 (Figure 3), indicating that these plants performed better in terms of FD, NFP, PH and BSD. In addition, this ecotype showed greater PH (46.5 %) and BSD (34.5 %) compared to the SAC ecotype (Table 5), and exhibited greater precocity (DF), coming 14 days earlier than the latest-flowering ecotype (LNL), and had 95 % more fruits per plant compared to the same ecotype (Table 5).

Figure 3 Graph of the principal components of the analyzed variables and ecotypes. PH = plant height; BSD = basal stem diameter; DF = days to flowering; AFY = average fruit yield; NFP = number of fruits per plant; FD = fruit diameter; NS = number of seeds; IL = internode length; NL = number of leaves; RTZ = Rio Tuxpan, Zacatecas; PTZ = Puente Tepetatilla, Zacatecas; LNL= Linares, Nuevo León; MZC = Múzquiz, Coahuila; SNL = Santiago, Nuevo León; SAC = San Alberto, Coahuila. 

On the other hand, blue netting increased the precocity to flowering by almost 18 days with respect to the open field, and, of all the covers, it also produced the largest IL, which was 62.81 % longer than in the production without cover. In this same variable, the RTZ ecotype was higher (27.38 %) than the SAC ecotype (Table 5).

The yield, size and quality in Capsicum, caused by the shading effect, depend largely on the geographical area and production technology (^{Zhu, Peng, Liang, Wu, & Hao, 2012}). In general, the RTZ and MZC ecotypes had the largest FD (40.59 and 20.48 %, respectively) and NS (46.52 and 46.04 %, respectively), compared to the SAC ecotype (Table 5). ^{Salinas-Hernández, Liévano-Liévano, Ulín-Montejo, Mercado, and Petit-Jiménez (2010)} also obtained greater length, FD and NS in two types of wild amashito pepper ("Pico de Paloma" and "Garbanzo"), when compared with other types of pepper evaluated.

^{Santos and Salame-Donoso (2012)} found that the use of 35 % shading in southern highbush blueberry (Vaccinium oxycoccus) cultivation increases flowering, fruit weight and yield, in contrast to open field production. On the other hand, ^{Nooryazdan, Serieys, Baciliéri, David, and Bervillé (2010)} report that differences in morphological traits, when evaluating 77 populations of wild Helianthus annuus collected in the United States, were correlated with the climatic variables of the sites of origin of the populations, which suggests adaptation to the local conditions of the evaluated species.

Effect of the interaction

Two-way analysis of variance revealed that the interaction between the covers used and the ecotypes had a significant effect on all variables. This coincides with the findings of ^{Hernández-Verdugo et al. (2015)}, who report significant differences in all phenotypic characteristics evaluated in wild pepper under different shade levels. In general terms, it can be pointed out that the use of white shading nets provides better results in the variables analyzed in all the ecotypes (Table 5), and that the RTZ ecotype presents the best values in all the covers used.

Analysis of the vectors revealed that a certain group of variables had a high correlation with each other (BSD, AFY, PH and NFP) (Figures 2 and 3). These results are similar to those reported by ^{Meena-Prakash and Bahadur (2015)} for PH and NFP, and by ^{Ogwulumba and Ugwuoke (2013)} for NL and NFP. On the other hand, the variables FD, IL, NS and NL formed another group, which showed little correlation with the previous group.

^{Meena-Prakash and Bahadur (2015)} indicate that PH showed a significant negative correlation with respect to fruit weight and polar diameter, which reveals that at a higher PH, fruit weight and diameter decrease, while the DF variable presented a negative correlation with all variables, which indicates that the higher the number of DF, the lower the performance of the other variables. In this study, biplot analysis revealed that all variables, except DF, are positively related to crops under white or blue nets (Figure 2); that is, crops under those covers had the best results in almost all variables. In terms of ecotypes, RTZ was positively related to the variables BSD, AFY, PH and NFP (Figure 3).

The DF variable in open field and greenhouse (Figure 2) reveals that plants grown under these conditions take longer to present flowers and have a lower performance in the rest of the variables. The same happened in all ecotypes when they were grown in the open field (Figure 3).