Growth Response of Four Conilon Coffee Varieties ( Coffea canephora Pierre ex A . Froehner ) to Different Shading Levels

In order to select a conilon coffee (Coffea canephora Pierre ex A. Froehner) adapted to shade, four varieties (C153, JM2, LB1 and GG) were submitted to four shade levels (0, 30, 50 and 70) with evaluation of plant height, stem diameter, collar diameter, number of plagiotropic branches, number of fruits, root colonization by arbuscular mycorrhizal fungi and leaf anatomy. The experiment was carried out on a completely randomized design, in a 4 × 4 factorial scheme with ten replicates. We used organic compost based on cocoa shells for plant nutrition and cow urine for phytopathogenic fungi control. There was interaction between variety and shade factors for most of coffee characteristics analyzed. In general all coffee characteristics evaluated mainly fruit number and length of fruiting branches, significantly increased with increasing shade. Coffee varieties tested respond differently to the increasing of shade levels and leaf anatomy demonstrated the reduction of mesophyll thickness as the shading increased. The effect of shade levels on fruit yield fit to a positive linear regression for all four coffee varieties tested but the mycorrhizal colonization no presented differences among coffee varieties and shade levels. The varieties C153 and GG presented highlighted anatomical, growth and productive characteristics and can be indicated for shading cultivation, for example on agroforestry systems.


Introduction
Brazil is the world's largest producer and exporter of coffee and the State of Bahia is the third largest producer of conilon coffee (Coffea canephora Pierre ex A. Froehner) on Brazil.According to data released in the annual survey of the coffee harvest conducted by CONAB, the main coffee conilon producing states in 2018 are: Espírito Santo, Rondônia and Bahia, which is responsible for approximately 13.66% of the national production of this coffee species (CONAB, 2018).conilon coffee.Shading can modify structural and functional characteristics of the leaves, influencing changes in the pattern of plant growth and production (Bjorkman, 1981).The plants submitted to a low luminosity condition develop anatomical changes, which play an important role in adapting the plant to the conditions imposed by the environment.These changes are related to increased uptake and utilization of incident light, a feature that limits shade growth, increasing the photosynthetic efficiency of the plant (Lambers et al., 1998).According to Castro et al. (2009) the leaf is one of the most plastic organs in response to environmental factors, and the variations resulting from these factors are observed in the foliar tissues in the attempt to adapt the plant.The knowledge of the leaf anatomy of the coffee tree is important to understand several physiological processes of the plant (Camargo & Marenco, 2011).
The study of the plant-microorganisms relationship have been intensified in the last decades, mainly due to the great importance that the microorganisms exert in the maintenance of the life, balance and restoration of ecosystems.The coffee tree has high mycorrhizal dependence, being this association favored mainly in low fertility soils (Lopes et al., 1983;Siqueira & Colozzi-Filho, 1986).The benefits that arbuscular mycorrhizal fungi (AMF) bring with the absorption of other nutrients have been well studied and they are efficient and of great importance for the plant nutrition (Moreira et al., 2010).
The evaluation of the type and magnitude of physiological, anatomical and morphological alterations that occur in plants submitted to reducing irradiation can contribute to the selection of more adapted and shade tolerating plant varieties.The objective of this research was to verify the effect of different shading levels on four different conilon coffee varieties evaluating the growth, yield and leaf anatomy, as well as AMF root colonization, in order to provide subsidies to technicians and coffee producers on the appropriate choice of variety suitable for cultivation on agroforestry systems.

Materials and Methods
The experiment was carried out in an experimental area located at the Universidade Estadual de Santa Cruz (UESC) located in the municipality of Ilhéus (Bahia State), from July 2016 to December 2017.The experiment site is located at 39°10′20.586″W, 14°47′47.769″S at 44 m above sea level.According to Köppen the climate classification is Af. with 24.4 °C of average temperature.The average annual rainfall is 1,419 mm (CLIMATE-DATA.ORG, 2017).
The experiment was a completely randomized design, on a 4 × 4 factorial scheme, the factors being four levels of shading (0%, 30%, 50% and 70%) and four varieties of conilon coffee (C153, JM2, LB1 and GG), with 10 replications per treatment.Coffee seedlings were transplanted into 20 L plastic pots filled with a 9:1 (v/v) mixture of soil and organic compost.The amount of compost mixed to the soil was established according to the K nutritional requirement of the crop for the first year of cultivation (60 mg dm -3 ), which corresponded to 2 kg of the organic compost.The same amount of compost was added to the soil 180 days after transplanting.
At 18 months after transplanting, when the coffee trees were on fruiting season, the following variables were evaluated: plant height, collar diameter, length of main branch, number of plagiotropic branches, number of fruits per plant, arbuscular mycorrhizal fungi (AMF) colonization of roots and number of AMF spores in soil.
For the anatomical evaluations of the coffee varieties, three completely expanded leaves of the third node of plagiotropic branches were collected and were cut in the middle third with razor blades and prepared according to Johansen (1940).The anatomical sections were made according to O'Brien et al. (1964).To determine the stomatal density, the technique of scanning electron microscopy was used.Measurements of the thickness of the     The incre mesophyll (Table 1).Note.Means values with the same lowercase letters in a row and with the same capital letters in a column are not significantly different by Tukey test (P > 0.05).
The difference among leaf thickness of non shaded and shaded conilon coffee plants can be observed on Figure 10.Also we noted that, regardless of variety, the studied conilon coffee presented dorsiventral mesophyll (Figure 10).The palisade parenchyma is dense and composed by elongated cells perpendicular arranged to adaxial surface of the leaf (Figures 9C, 9D and 10).The spongy parenchyma presented large intercellular spaces (Figures 9C, 9D and 10).
The decrease on leaf blade thickness of the shaded conilon coffee probably occurred due to the leaf blade expansion and difference on photoassimilates distribution in relation to full sun cultivated plants, which presented thicker leaves.Our results corroborate those of Voltan et al. (1992) and Cordeiro-Gomes (2008) who verified that leaves of coffee trees exposed to higher levels of radiation had a greater thickness of the leaf blade.This increase in mesophyll thickness causes reduction of the specific leaf area, which, according to some authors, propitiates to full sun plants a greater photosynthetic capacity per unit of leaf area when compared to shaded plants (Niinemets et al., 1998).

Conclus
The four c demonstra the same growth and

Table 1 .
Leaf, mesophyll and parenchyma thickness (μm) of four conilon coffee varieties submitted to four different shade levels