Objectives

The aim of this study was to investigate whether the planting density has an influence on anatomical elements of wood from Eucalyptus urophylla × E. grandis and Acacia mangium.

Characterization of the study area

The experiment was established in January 2009 at the Instituto de Ciências Agrárias of the Federal University of Minas Gerais, in Montes Claros city, Minas Gerais, Brazil (16°40'03.60"S; 43°50'41.52"O; 598 meters above sea level) as described by ^{Silva et al. (2018)}. The climate is Aw-Tropical Savanna, characterized by high annual temperatures and rainy regime marked by two distinct seasons, rainy summer and dry winter (^{Köppen and Geiger, 1928}). The soil is characterized as a eutrophic red-yellow argisoil (^{Claessen, Barreto, de Paula and Duarte, 2006}).

Experimental design and sampling

The trees were planted in high density and low density stand systems with Eucalyptus urophylla x E. grandis hybrid (hereafter Eucalyptus) and Acacia mangium Willd (hereafter Acacia) as shown in figure 1. The plantation density in the monoculture system (high density) was 2 m × 3 m, while in the mixed stand (low density) the spacing was 10 m × 2 m. The experiment was conducted in a completely randomized design composed by four treatments, represented in this study by five replications. The four treatments were considered as: i) Eucalyptus in high planting density (monospecific plantations, 1667 trees/ha), ii) Eucalyptus in low planting density (mixed stands, 500 trees/ha), iii) Acacia in high planting density (monospecific plantations, 1667 trees/ha), iv) Acacia in low planting density (mixed stands, 500 trees/ha). Ten trees of Acacia and ten of Eucalyptus were selected (5 of each type of planting density) and harvested in August 2013, when the trees achieved four years and six months of age. The selection of these 20 individuals (2 species × 2 planting density × 5 repetitions) was based on the uprightness of the stem and absence of visible diseases.

Figure 1 Plantation scheme showing high and low density stands with Eucalyptus (white circles) and Acacia (gray circles) trees. 

In regard to the growing characteristics of trees, Eucalyptus presented diameter at breast height (DBH) of 18.92 cm ± 0.93 cm in monospecific plantations and 15.01 cm ± 1.56 cm in mixed stands, while Acacia had DBH of 14.64 cm ± 1.73 cm (monospecific plantation) and 11.11 cm ± 1.09 cm (mixed stands), as reported by ^{Silva, Roldão, Santos, Hein (2018)}.

Wood discs (thickness: 30 mm) were cut at breast height, properly identified, placed in plastic bags to keep moisture and then divided into four wedges, free from defects. Sixty-eight cubic specimens measuring 20 mm × 20 mm × 20 mm used in this study were cut from the wedges. For each wedge, cubic specimens were cut near the pith, at intermediate radial position and near the cambium.

Anatomical characterization of wood

Histological sections were made from the specimens in order to study the variation in the anatomical features of the wood. For preparing the slides, the wood specimens were immersed in water for 24 hours and then placed in a pressure cooker with water and glycerin during 15 minutes, to soften the wood.

The sections were produced using a sliding microtome with nominal thickness of 30 microns and semi-permanent sections were stored with glycerin. The thin sections were placed in bleach until they become clear and subsequently in distilled water, to remove traces of bleach. The sections were stained with Safranin and passed through an alcoholic series, from lower concentration (20%) to absolute ethanol, in order to remove the dye excess and dehydrate them, as described by ^{Johansen (1945)}. The slides were mounted with the thin sections, fixed with Entelan and covered with coverslips.

For measuring fiber diameter, wall thickness and fibers length, small fragments of wood were removed from specimens, placed in macerate solution of hydrogen peroxide and 1 N acetic acid at a ratio of 1:1 and conditioned in an oven at temperature of 60 °C for 48 hours, in order to promote fiber individualization, as suggested by ^{Franklin (1945)}. After this period, the fibers were washed in water, stained with diluted safranin and mounted in histological slides. The temporary mounts were prepared for the measurements on optical microscope equipped with an ocular micrometer and a graduated lamina with 1 mm of scale. The average fiber length and diameter was based on the observation of 30 fibers per lamina using a light microscope with10x objective lens. The cell wall thickness was measured by observing 30 fibers per sample from the macerated material through the 40x objective lens, as experimental procedure adopted by ^{Monteiro et al. (2017)}.

To measure the frequency of vessels, the mounts were analyzed using a 10x objective lens. Vessel diameter was observed and accounted for twin or multiple vessels; each unit was individually measured in tangential and radial direction and the average values were calculated for each unit, according to an adaptation of ^{International Association of Wood Anatomy [IAWA] (1989)}. The vessels were randomized including different sizes found in the sections.

The capture of images was performed using a digital camera coupled to an optical microscope (Ken-Vision, TT-1010 model) and the software WinCellPro was used for the anatomical measurements of the wood sections, which followed the IAWA (1989) procedure. The following quantitative anatomical features were evaluated: vessel lumen diameter (VD), vessel frequency (VF), fiber length (FL), fiber width (FW), fiber lumen diameter (LD) and fiber cell wall thickness (CWT). Each slide was composed by three replicates of cross section. Thirty measurements per slide were made for each anatomical feature.

Analysis of results

The statistical software "SPSS v.19" was used to calculate the descriptive statistics, analysis of variance (Oneway), multiple comparison among means (Tukey) and t-test. For the analysis of variance were considered four variation sources: 1) high density plantation of Eucalyptus hybrid; 2) high density plantation of Acacia mangium; 3) low planting density stands of Eucalyptus hybrid and 4) high planting density stands of Acacia mangium. The normality of the residuals and the homogeneity of variances were tested and the averages compared by Tukey test at 5% significance level.

Anatomical variability

The mean, values of the anatomical features observed in Eucalyptus grandis x E. urophylla hybrid and Acacia mangium grown in high and low density stands are shown in table 1. The Eucalyptus trees grown in low or high plantation density did not present significant differences between values of fiber width (FW), fiber lumen diameter (LD) and cell wall thickness (CWT). However, there was a significant effect of planting density on vessel diameter (VD), frequency (VF) and fiber length (FL) of Eucalyptus wood cells.

Acacia trees grown in low density stand (spacing: 10 m × 2 m) presented higher values in most anatomical parameters, but the CWT and VF were higher when the Acacia was grown in high density plantations.

Effect of planting density on vessels diameter and frequency

The two species produced wood with larger vessel diameters when cultivated in low density plantations (Table 1). The diameter of vessels found in this study for 4.5 years-old Eucalyptus wood in low density stand (125.9µm) was significantly higher than in high density plantation (120.1µm). The vessel diameters of these woods were slightly higher than those reported by ^{Alzate (2004)}, who found an average vessel diameter of 103.63 µm for Eucalyptus grandis x E. urophylla with 8 years old, and ^{Lima de Oliveira et al. (2012)} who found average values of 99.21 µm in several species of Eucalyptus with 64 months of age. Here, the average vessel diameter of Acacia wood was 133.7 µm and 119.2 µm for the low density stand and high density (Table 1), respectively (the VD of Acacia wood were significantly different by t-test). These values are consistent with those reported by ^{Antunes (2009)}, who found vessels with diameters ranging from 120 microns to 160 microns in Acacia crassicarpa and Acacia mangium wood.

The variation in vessel frequency of the Eucalyptus and Acacia wood grown in high and in low density stands are shown in figure 2.

Figure 2 Variation in vessel frequency of Eucalyptus grandis x E. urophylla and Acacia mangium grown in high and low density stands. 

In regard to the vessels frequency, there was a significant difference between treatments (Table 1). Figure 2 reveals there was effect within each density system. Eucalyptus wood planted in low density system produces significantly more vessels per unit area than the wood produced in high density. In Acacia plantations, the reverse trend was observed.

The frequency of vessels observed by ^{Alzate (2004)} was of 8-13 vessels/mm² in 8-year-old Eucalyptus grandis x E. urophylla hybrids. ^{Monteiro et al. (2017)} reported frequency of 16 vessels per mm² and ^{Evangelista, Silva, Valle and Xavier (2010)} found 9.9 vessels/mm² in E. urophylla wood. ^{Antunes (2009)} reported vessels frequency from 4 to 9 vessels/mm²in A. mangium wood.

^{Sousa Junior (2004)} has evaluated the differences in diameter and frequency of vessels from E. urophylla wood grown in distinct spacing plantations (6 m × 6 m and 3 m × 2 m), reporting lower vessel frequency (10,89 mm^-²) and higher vessel diameter (105,85 µm) in the wood produced in lower planting density than the wood from the higher planting density (VF of 11,37 mm^-², VD of 96,09 µm). However, the trees growing at different planting density had different ages and provenance: trees planted at 6 m × 6 m had 15 years old and came from Paraopeba city, while the other trees had 25 years old and were cultivated at Turmalina city, Brazil. To our knowledge, the study conducted by Sousa Junior (2004) is the closer one to our findings on Eucalytptus, but it is difficult to make comparisons because of such differences in provenance and age.

Fiber biometric changes according to planting density

The results shown in Table 1 indicate that both Eucalyptus and Acacia trees planted in low density system produce significantly longer fibers. This finding agrees with those found by ^{Teago (2012)}, who studied Acacia mangium and Eucalyptus woods and found that in monoculture these trees produce wood with shorter fiber, besides having greater radial heterogeneity for the fiber width, lumen diameter and wall thickness of the fiber.

The length of Eucalyptus fibers found in this study are in agreement with the literature (^{Alzate, 2004}; ^{Alencar, Barrichelo and Silva 2002}; ^{Brisolal and Demarco 2011}). But the Acacia fibers are longer than those reported in similar studies. ^{Alencar et al. (2002)} reported fibers 1.18 mm long in 4 year-old E. grandis x E. urophylla wood while ^{Brisolal and Demarco (2011)} analyzed 6 year-old E. grandis x E. urophylla with fiber length of 1.1 mm. ^{Tomazello-Filho (1983)} reported 10 year-old Eucalyptus grandis with fiber length of 1.03 mm. Regarding the fiber length of Acacia wood, ^{Alencar et al. (2002)} found fibers of 0.89 mm and ^{Rossi, Azevedo and Souza (2003)} found fibers measuring from 1.0 mm to 1.2 mm, a little longer than the fibers of the wood investigated in this study.

The fibers width does not seem to be sensitive to planting space. Table 1 shows no significant difference between the planting density for both species. However, there was a species effect on the variation of this fiber feature. The Acacia wood has wider fibers (~ 20 µm) than Eucalyptus (~ 16 µm). The fibers width of the E. grandis x E. urophylla hybrid is in agreement with the literature: ^{Alencar et al. (2002)} found fibers with 19.95 µm in width in 4-year-old E. grandis x E. urophylla; ^{Brisolal and Demarco (2011)} reported fibers of 20 µm in width and ^{Alzate (2004)} evaluated wide fibers, with average of 19.7µm. In Acacia, the literature indicates a larger variation between fiber widths: 24.7 µm in ^{Alencar et al. (2002)} and 16 mm in ^{Antunes (2009)}.

The lumen diameter of Eucalyptus fiber was not influenced by planting density while there was a significant difference between the mean lumen diameters of the A. mangium samples. The Acacia mangium wood produced in low density stand had fibers with bigger lumen diameter when compared to fibers produced in high density plantation (Table 1). This characteristic is within the range found in the literature for Eucalyptus (^{Alencar et al., 2002}; ^{Brisolal and Demarco, 2011}; ^{Alzate, 2004}) and Acacia (^{Alencar et al., 2002}; ^{Antunes, 2009}).

There was no variation in cell wall thickness of the fibers of Eucalyptus trees grown in high and low density stands. However, when Acacia is planted in high density, the cell wall thickness of the fibers is higher than when it grows in low density stands (Table 1).

The cell wall thickness varies greatly in Eucalyptus literature: 4.31 mm in ^{Alencar et al. (2002)}; 6.1 µm in ^{Brisolal and Demarco (2011)}; and 5.01 µm in ^{Alzate (2004}). The cell wall thickness of the Eucalyptus wood in this study had lower values (~3.2 µm). For Acacia, the cell wall thicknesses found in this study (3.1 µm - 3.4 µm) were close to the literature: 3.59 µm in ^{Alencar et al. (2002)} and 3.2 µm in ^{Antunes (2009)}. Cell wall thickness is important because it is highly correlated to the density of the wood and also influences its hardness (^{Cutler, Botha and Stevenson (2008)}.