WOOD AND CHARCOAL ANATOMY OF FOUR MYRTACEAE SPECIES

Qualitative and quantitative changes in anatomical characteristics of charcoal from Campomanesia xanthocarpa, Eugenia pyriformis, Myrcia retorta and Plinia peruviana were described to provide its correct discrimination and information’s to contribute to evaluations in illegal logging. The material came from Planalto Catarinense region, southern Brazil. For each species, three discs at diameter at breast height were collected from three randomly selected trees. All the disks were sectioned to obtain samples in different positions of the trunk (bark, intermediate and pith). Samples were wrapped in aluminum foil and carbonized in a muffl e furnace, with a fi nal temperature of 450 °C and a heating rate of 1.66 °C.min-1. In species analyzed, growth rings are few distinct, delimitated by fi ber regions; solitary vessel and axial parenchyma diffuse or diffuse-in-aggregate is predominant. The wood to charcoal conversion caused variation in the following characteristics: vessel diameters decreased varying from 21.1% in Myrcia retorta to 33.4% in Eugenia pyriformis, while the vessel frequency increased with values from 50.7% to 92.3% related to the same species; height and width of the rays showed reduction, however, Myrcia retorta presented ruptures in ray cells and dimensions increased; ray frequency increased from 3.9% in Eugenia pyriformis to 54.1% in Campomanesia xantocharpa. Qualitative characteristics of wood from studied species remained in charcoal after carbonization at 450 oC, being possible to discriminate the material based on wood anatomy. In charcoal, vessels dimensions and frequency, after ray characteristics can be applied as fi rst step in this Myrtaceae species distinction at forest control. v.24 n.3 2018


INTRODUCTION
In tropical forests of world, in charcoal production, literature report the use of native species from deforestation regions, and also that there is no selection of cut species. The process impact soil, affects global warming through the production and emission of gases, such as carbon dioxide and methane (Chidumayo and Gumbo, 2013). In Brazil, statistics shows that this scenario is being modified in function of environment legislation and more control in inspection process. In 2016, charcoal manufacture from native forest present a reduction of 31.7% (IBGE, 2016) and the use of planted species is estimated in 84% of total consume of country (IBÁ, 2017). On the other hand, data no included in official statistics shows that charcoal production present a high economic and social importance for much family farmers in southern Brazil and, in this case, native species are applied (Bauer et al., 2015).
In Santa Catarina, Brazil, the Araucaria Forest is composed by 925 species from 439 genera and 116 botanic families. Myrtaceae family is the second in number of species related (Gasper et al., 2013) and some of them are listed as endangered on Brazilian legislation, for example 5 species from genera Campomanesia, 20 from Eugenia, 10 from Myrcia and 8 from Plinia (MMA, 2014). A correct identification of species is important in forest control, and as in charcoal qualitative anatomical characteristics remain (Braadbaart and Pole, 2008;Kwon et al., 2009), a database from different species can help in its evaluation.
Changes in cell dimensions after carbonization process are described in literature as a response of anatomical and chemical composition of wood species and also process characteristics (heating rate, final temperature and total time), but qualitative information remains and can be applied in species identification (Nisgoski et al., 2014;Muñiz et al., 2016;Gasson et al., 2017). In general, tangential diameter of vessel presents reduction and its frequency an increase (Gonçalves et al., 2012 but are not linear and are related to species characteristics as fiber cell wall and parenchyma distribution (Gonçalves and Scheel-Ybert, 2016).
Some studies with charcoal from native species are present in literature, but from Myrtaceae family only charcoal from Myrcia bella  was found. So, considering charcoal production and commerce with native species in southern Brazil, this paper has the objective to evaluate and compare wood and charcoal from four Myrtaceae species, to provide a correct discrimination and contribute to control of illegal logging, by providing information on carbonized wood.  For each tree, a disc at diameter at breast height (DBH) was selected for evaluations and a wedge with no defects of each one was cut. Material was divided into three samples oriented in anatomical sections (transversal, radial and tangential), with 2 x 2 x 2 cm, named as: near pith, intermediate, and near bark. For wood macroscopic characterization, samples were sanded with 2000 sandpaper.

MATERIAL AND METHODS
Wood and charcoal analysis was made in the same samples. After wood evaluation, all samples (moisture content of 12 ± 1%) were wrapped in aluminum foil and carbonized in a muffle furnace, with a final temperature of 450 °C and a heating rate of 1.66 °C . min -1 based on Muñiz et al. (2012a). The carbonized material remained at the final temperature for two hours. Macroscopic description of wood and charcoal was based on images obtained in a Stereomicroscope Discovery V12 (Zeiss) with software Axio Vision Rel. 4.7. Charcoal details were obtained in a Hitachi TM-1000 tabletop scanning electron microscope (SEM) directly from the material, without coating. The description of the anatomical elements of CERNE STANGE et al. wood and charcoal samples followed the orientations of the International Association of Wood Anatomists (IAWA, 1989). Number of measurements were adapted on the basis of 30 readings regarding frequency and tangential diameter of the vessels; frequency, height and width of the rays in micrometers.
Data analysis was done in a factorial 2 x 3 considering kind of material (wood and charcoal) and different positions in disk (near bark, intermediate, near pith). Differences in dimensions between wood and charcoal were evaluated by Scott-Knott test at 95% of probability.

Myrcia retorta Cambess
The wood present: Growth rings: few distinct marked by fibers regions. Vessels: diffuse-porous ( Figure  3A), rounded, solitary predominant, simple perforate plate, intervessel pits alternate. Axial parenchyma: apotracheal diffuse and diffuse in aggregate, scanty paratracheal. Rays: heterogeneous, body ray cells procumbent with rows of upright and/or square marginal cells; multiseriate, not storied. Fibers: libriform, nonseptate and thin-walled. In charcoal, some changes were verified after carbonization process. In transversal section it was possible to verify cracks in rays, probably in function of thin-walled fibers. Vessel present retraction ( Figure  3B,C); few contrast in axial parenchyma; heterogeneous rays ( Figure 3D) and multisseriate rays ( Figure 3E) are more evident in charcoal. Simple perforate plate is also visible in radial section ( Figure 3F).

Quantitative characteristics -variation in cell dimensions
Tangential diameter and frequency of vessel Means of vessel diameter (Table 2)    For Eugenia pyriformis, all positions presented decrease in vessel diameter after carbonization process, varying from approximately 32% in region near bark and intermediate, and 37% in region near pith. Also species shows a gradual reduction from bark to pith in vessel diameter of wood and charcoal. In wood of other species from genus Eugenia, Santos and Marchiori (2011) reported Observing Plinia peruviana wood, intermediate region present vessels with larger diameter, while region near bark and pith are similar. After carbonization process all samples showed reduction in vessel diameter, a reduction of 23%, 36% and 29% was verified for regions near bark, intermediate and near pith, respectively. Comparing only charcoal dimensions, region near bark is different from others and present vessels with bigger diameter. Small vessel diameter in genus Plinia is also reported for other species, for example, Plinia rivularis present values from 32.5-70 µm and Plinia trunciflora from 17.5-38.7 (Santos and Marchiori, 2011). All positions of wood are different in number of vessel, and fewer vessels were observed in intermediate region. In charcoal, intermediate region is different from near bark and near pith. In all samples an increase after carbonization process in vessel frequency was verified with little variation between positions: 86% in region near bark, 84% in intermediate region, and 82% near pith. In wood from other species from genus Plinia, a frequency of 75-162 was reported in literature (Santos and Marchiori, 2014).
Reduction in vessel diameter after carbonization process is the result of material contraction and must be influenced by species characteristics, and process conditions as heating rate, the difference in regimes and the duration of the process (Nisgoski et al., 2014). In carbonization process, chemical components of cell wall are degraded, and in consequence, the volume occupied by charcoal in a furnace is less than wood volume inside. This fact characterizes contraction of carbonized material and reduction in vessel diameter (Pereira et al., 2016). Higher extractives contents associated with lower crystallinity and lower crystallite size can accelerate the degradation process and reduce wood thermal stability (Poletto et al., 2012). Also fiber cell wall and kind of axial parenchyma can influence dimensional changes in vessel diameter (Muñiz et al. 2012aPereira et al., 2016).
In this study, evaluating mean reduction of vessel diameter by species, it was observed a great decrease in Eugenia pyriformis (33.4%), in sequence, Plinia peruviana (29.4%), Campomanesia xanthocarpa (21.7%) and Myrcia retorta (21.1%). Some studies in native species report reduction in vessel diameter after carbonization process, for example, in Cerrado species Gonçalves et al. (2012) related decrease in 70% of evaluated material; in five wood similar to angelim, Muñiz et al. (2016) verified changes in all species studied; and in five species from Santa Catarina state, Araucaria Forest, Carvalho et al. (2017) reported decrease in tangential diameter of vessels in all samples.
Related to variations of vessel frequency by species after carbonization process, it was observed a great increase in Eugenia pyriformis (92.3%), in sequence, Plinia peruviana (84.0%), Campomanesia xanthocarpa (60.3%) and Myrcia retorta (50.7%). In all species it was observed increase in frequency and decrease in vessel diameter (Table 2) after carbonization and is the result of cell contraction in function of temperature. Changes in vessel frequency after carbonization process were commented in other studies: Nisgoski et al. (2014) evaluating samples of Ocotea porosa in different final temperatures in carbonization process observed no direct relation between decrease in vessel diameter and increase in vessel frequency. Other studies with forest species (Muñiz et al., 2012b;Nisgoski et al., 2012), angelim species  and Anacardiaceae species (Gonçalves and Scheel-Ybert, 2016) also showed the influence of species characteristics in changes in vessel frequency, as thickness of fiber wall and disposition of parenchyma cells.

Ray height and width
In most samples it was observed changes in ray height (Table 3). A reduction was present in Campomanesia xanthocarpa, Eugenia pyriformis and Plinia peruviana, and an increase in ray height occurs in Myrcia retorta. Mean value of ray width shows different comportment between position and species and also after carbonization process (Table 3). Differences in species can be related to percentage of multiseriate and uniseriate rays.
In ray height, in Campomanesia xanthocarpa wood, it was not observed differences between positions in direction pith to bark. In charcoal, region near bark present ray with the tallest dimensions. Decrease in ray height varied from 11% (near bark) to 18% (intermediate and near pith). Means results in wood are similar to values reported in samples from Atlantic Forest by Santos and Marchiori (2011) which related a mean ray height of 277 µm (140-570), and Soffiatti et al. (2016) who verified mean value of approximately 246.7 µm (51.35-480.86). In wood and charcoal it is possible to notice an increase in ray width from pith to bark. After carbonization, changes occurred in intermediate and region near bark, with decrease in ray width of 6% and 12%, respectively. In region near pith 5% of reduction was verified, with no statistical differences. Mean values of wood are greater than observed in tress from Araucaria Forest in Paraná state, which varied from 5.7 to 41.8, with mean of 26.3 µm described by Soffiatti et al. (2016).
In Eugenia pyriformis wood and charcoal, ray height is similar in region near bark and intermediate portion, being greater than near pith. In all positions a decrease in dimension after carbonization process was observed with some different change, varying from 17% (near bark) to 20% (near pith), being 18% in intermediate region. Different species of Eugenia from Rio Grande do Sul, described in literature present ray with height varying from 92-1100 µm, showing a great difference (Santos and Marchiori, 2011). Some differences in positions were verified for wood and charcoal related to ray width. In wood, region near pith presented the smallest ray width and is statistically diverse from other positions. In charcoal, region near bark present the biggest dimension and is different from regions intermediate and near pith. Charcoal showed and decrease in ray width of 30% in region near bark, of 34% in intermediate position, and of 29% in region near pith.
In the species Myrcia retorta wood there is no difference between samples position, and in charcoal region near pith shows smallest rays, statistically diverse from region near bark and intermediate. Samples present diverse results from other species, after carbonization process an increase in ray height in all positions was observed, varying from 16% near pith to 34% in intermediate region. These results are related to ruptures in rays that were the result of expansion of cells walls. Ruptures in rays are also reported in Quercus sp. (Kim and Hanna, 2006), in some eucalyptus species (Gonçalves et al., 2014) and eucalyptus clones (Pereira et al., 2016). In wood, values are smallest than from Myrcia hatschbachii which present mean ray height of approximately 418 µm, varying from 172-899 µm (Soffiatti et al., 2016). In wood, ray width presented some differences in region bark, and for charcoal it was not observed variations in positions from pith to bark. All samples showed an increase in ray width after carbonization process, and changes of more than 50%. In wood, values of studied species are smallest than from Myrcia hatschbachii which present mean ray width of approximately 22 µm (Soffiatti et al., 2016). approximately 6% of reduction. The influence of sample position in ray width was observed only in region near pith, in wood. In charcoal, all material was different. After carbonization process, a decrease in all samples occurred. Changes varied from 26% (region near bark) to 34% (intermediate region).
In this study, evaluating mean changes of ray height by species, it was observed an increase of 24.3% in Myrcia retorta, and a decrease, in sequence, for Eugenia pyriformis (18.3%), Campomanesia xanthocarpa (15.7%) and Plinia peruviana (10.0%). Other studies with native species also report different comportment, in the same carbonization process of this study, related to thickness of fiber wall (Muñiz et al., 2012a), to characteristics of ray cell composition (homogeneous or heterogeneous) and comments about specific variations intra and inter individuals (Gonçalves et al., 2012).
In this study, evaluating mean changes of ray width by species, it was observed an increase of 56.3% in Myrcia retorta, and a similar decrease for Eugenia pyriformis and Plinia peruviana (30.7%), and few reductions in Campomanesia xanthocarpa (7.6%). Decrease in ray width was also reported in Enterolobium schomburgkii (Muñiz et al., 2012a). The same behavior was related by Gonçalves et al. (2012) in Stryphnodendron polyphyllum, Croton grandiflora and Vochysia tucanorum. On the other hand, an increase in ray width was also observed in Cedrelinga catenaeformis (Muñiz et al., 2012a) and can be related to chemical changes in substances stored at parenchyma cells which can expand and result in gas components.
Variations in ray width by species can explain different comportment. It was commented by Kim and Hanna (2006), evaluating Quercus variabilis, that in uniseriate ray its morphological characteristics remained and cell presented small separations, instead of multisseriate rays showed disintegration in cell wall and honeycombed wood was verified with increase in carbonization temperature.

Ray frequency
After carbonization different behavior was observed and influence of position and species was verified (Table 4). Variations in position of Campomanesia xanthocarpa wood was only present between region near pith and other positions. In charcoal, all positions showed difference in ray frequency. After carbonization process an increase in ray frequency was observed for all positions, varying from approximately 52%, in region near pith and near bark, to 58% in intermediate region.
In wood, Santos and Marchiori (2011) verified a mean ray frequency of 13 mm -1 , varying from 10-15 evaluating wood from Rio Grande do Sul.
In Eugenia pyriformis wood, region near bark present less vessels by millimeter being different from other positions. In charcoal it was observed more rays in region near pith, being diverse from other samples. Changes in ray frequency after carbonization occurred only in position near bark, with an increase of 6%.
For Myrcia retorta positions from pith to bark present no influence in ray frequency in wood and charcoal. Some increase in frequency was observed for region near bark (10%) and intermediate region (7%).  reported in charcoal frequency of 17, varying from 13-21 rays by millimeter in Myrcia bella, in wood Sonsin et al. (2014) verified a mean value of 13 mm -1 .
Influence of position in samples from Plinia peruviana was only observed in intermediate region of wood material, which resulted in less frequency and is different from others. In charcoal production, an increase in ray frequency was verified for all positions, varying from approximately 29% in region near pith to 57% in intermediate region.
In this study, evaluating changes in ray frequency by species, it was observed an increase of 54.1% in Campomanesia xanthocarpa, 41.3% in Plinia peruviana, 5.9% in Myrcia retorta and 3.9% in Eugenia pyriformis. These variations are the results of different contraction and homogenization of cell wall. Ávila et al. (2017) comments that these variations can be explained by mass loss and wood contraction in function of carbonization process, but also intraspecific and ecological trends can be involved.

CONCLUSION
The wood to charcoal conversion caused variation in the following characteristics: decrease in vessel diameter; increase in vessel frequency; decrease in ray height and width in Campomanesia xanthocarpa, Eugenia pyriformis and Plinia peruviana and opposite behavior in Myrcia retorta; increase in vessel frequency in all species. The species Myrcia retorta presented ruptures in ray cells.
Qualitative characteristics of wood from studied species remained in charcoal after carbonization at 450 ºC, being possible to discriminate the material based on wood anatomy. In charcoal, vessels dimensions and frequency, after ray characteristics can be applied as first step in this Myrtaceae species distinction at forest control.