Anatomical Characteristics of Garcinia lucida (Vesque) and Scorodophloeus zenkeri (Harms) Wood and Debarking Response in the South Region Cameroon

The aim of this work was to investigate the possible anatomical changes of Garcinia lucida and Scorodophloeus zenkeri after the removal of their bark. Debarking was done on individuals of each species at 1.30 m from the soil. The wound was rectangular in shape with 30 cm side. There was a follow-up every three months for nine months during which the survival and rate of regeneration of the bark were recorded. A block of cube was cut from the regenerated and intact wood of species for microtomy and microscopy activities. On the cross-section of each wood, vessel features like density and diameter were measured before and after wounding. Semi-automatic measurements were made using the SpectrumSee digital image analysis software. In the wood of the two species, it appeared that the density of the vessels before debarking was significantly comparable to the density after debarking, while the diameter of vessels in the regenerated wood was smaller. The cambial area increased slightly in the rainy season for all species. After nine months all the species started the restoration of their conductive zone. G. lucida heals its wound more rapidly than S. zenkeri.


Introduction
Tropical forests are ecosystems which suffer from the depletion of their surfaces and this is directly linked to population growth (Millet, 2001). Thus, the importance of non-wood forest products (NWFP) for local livelihoods and as a means to ensure forest conservation has been widely recognized (Marshall, Schreckenberg, & Newton, 2006). The exploitation of NWFPs has become an increasingly attractive activity for many poor and unemployed people in Central Africa in general and Cameroon in particular (Guedje, Lejoly, Nkongmeneck, & Jonkers). NWFPs are defined as goods of biological origin other than wood, derived from forests, wooded lands and trees outside forests (Food and Agriculture Organization of the United Nation, 1999). About 80% of the people in developing world depend on NWFPs for their primary health and nutritional needs and/or in terms of meeting their subsistence consumption and income needs. Sustainable forest management requires that management ensures the utilization that meets present demand for forest products and services and also ensures that future generations' ability to have the products and services is not compromised. Furthermore the forests are able to regenerate and maintain existing condition (Chamberlain, Bush, & Hammett, 1998). The exploitation of tree species has a variable effect depending on the parts harvested since harvesting of the bark is more damaging in terms of tree survival (Geldenhuys, Syampungani, Meke, & Vermeulen, 2007;Vermeulen, 2009). The bark is produced by a thin layer of cambium cells which surround the xylem and phloem tissues that transport water and nutrients to and from the roots and leaves. The bark also protects plants against fire, fungal and insect attack (Cunningham, 2001). Removal of bark can therefore damage phloem or expose it to desiccation and fungal or parasite attack. This may disrupt the conduction of nutrients and hormones involved in flower bud production for example (Mohr & Schopfer, 1995). The need for resources to repair damage to bark could also result in lower resource allocation to reproduction. Trees respond differently to harvesting methods, not much information is available on the ecological impacts of bark harvesting (Ticktin, 2004). Bark harvesting interrupts suddenly the water relationship between bark and wood and may affect the water conduction between leaves and roots (Zwieniecki, Melcher, Field, & Holbrook, 2004). As trees consume large amounts of water, they have to develop mechanisms for protection against disturbance of their water balance whereas they should be able to restore the water pathway. Both the diameter and the density of vessels directly influence conductivity (Christensen-Dalsgaard, Fournier, Ennos, & Barfod, 2007;Sellin, Rohejarv, & Rahi, 2008).
In southern Cameroon, Garcinia lucida (Clusiaceae) is harvested from undisturbed and generally easy to access forests. Moreover, this species may be endangered by the overexploitation caused by the use of its bark in the production of palm wine. The bark is used as an additive to palm wine production and in distilling fermented palm wine to produce liquor. The bark and the seeds, dried or fresh, are widely used for medicinal purposes to prevent food poisoning and to cure stomach and gynaecological pains, as well as to cure snake bites (Guedje, Tadjouteu, Onana, Nnanga Nga, & Ndoye, 2017). This species is a gregarious species that grows in dense stands on the slopes of forests above 500 meters of altitude. However, there are several sites with a relatively high mortality rate, resulting from an anarchic and intensive exploitation of this resource (Guedje, Zuidema, During, Foahom, & Lejoly, 2007). Studies on the exploitation and management of Scorodophloeus zenkeri in the Ngovayang forest in southern Cameroon reveal that trees are generally cut down when diameter at breast height is more than 35cm for the exploitation and marketing of their bark which is used as a condiment or spice (Ndzomo, 2011).
The aim of this paper was to evaluate the short-term responses to wounding in two tree species. We are going to describe first the wood of the two species following IAWA characteristics. Since vessel features can be considered as indicators of anatomical wood reactions following bark harvesting, we should investigate the impact of the wood after nine months by evaluating the features of vessels and other wood elements produced before and after bark harvesting.

Location of Study Area
The Ngovayang massif, in southern Cameroon, is part of a group of small hills along the Atlantic coast of the Gulf of Guinea. This area is known for its floristic richness and its high level of endemism (Gonmadje, Doumenge, Sunderland, Balinga, & Sonké , 2012). It extends over 102 000 ha between the parallels 3 ° 12'-3 ° 25'N and the meridians 10 ° 30'-10 ° 45'E. The altitude varies between 50 m in the south-western part and 1090 m in the central-eastern part, on the highest peaks. The climate is of the sub-equatorial type (Suchel, 1972). The average annual precipitation is around 2000 mm (Waterloo, Ntonga, Dolman, & Ayangma, 2000). They decrease with increasing distance from the ocean. The average annual temperature is 25 °C (Waterloo et al., 2000). The dense river system is made up of numerous small rivers intersected by falls. Biafrean Atlantic forests rich in Leguminosae-Caesalpinioideae constitute the main type of vegetation encountered in the study area. In addition to global threats such as climate change, destruction of the forest for arable land use and logging have contributed to the destruction of natural vegetation for decades, mainly in low-lying and flat areas that are easily accessible (Letouzey, 1985). Other human pressures, such as hunting or harvesting of forest products, are also notable, even at higher altitudes (Gonmadje et al. 2012).

Description of Study Species
Garcinia lucida species is a small understory dioecious tree, standing sometimes on stilt roots, reaching 25 -30 cm in diameter at breast height (dbh) and 12-15 m in height. Trees are of variable aspect, well-branched and evergreen. The stem exhibits yellow and resinous exudates after the dark brown bark is cut or removed. The natural habitat of G. lucida is the Atlantic "primary" rain forests. G. lucida tree grows in high-density stands covering few hectares with an abundance varying from 348 young matures (5 -10 cm dbh) to 5 large trees (> 20 cm dbh) per hectare (Guedje et al. 2017).
Scorodophloeus zenkeri (Caesalpiniaceae) is a tree up to 35 m tall and 200 cm in diameter. Its trunk is straight and more or less cylindrical. The base is slightly thickened and has grooves. The top of this tree is open with upright branches. The bark of Scorodophloeus zenkeri is gray-yellow, smooth in young individuals then scaly and may have a slice about 1 cm thick, brittle, yellow with a garlic odor (Eyog, Ndoye, Kengue, & Awono, 2006).

Sampling of Bark Recovery Parameters
For each species, the experimental debarking was carried out on 11 individuals of a diameter class between 10 and 30 cm, during the dry season. These individuals were chosen randomly, five individuals for S. zenkeri and six for G. lucida. For each species, only trees that had not yet been debarked were selected for the experiment. On each individual, wounds were made at 1 m above ground level. The wound consisted of a rectangular piece of bark 30cm vertically and 10cm width. Six months after, block of wood and bark samples were cut at wound level, in the intact zone and regenerated zone and were immersed in 1/3 ethanol (diluted to 70%), 1/3 glycerin and 1/3 water so as to preserve the intact cambial zone and to ensure the good wood-bark cohesion at least for 72 hours.
Samples sections (transversal, tangential and radial) with a thickness of about 15-30μm were obtained and contain phloem, cambial zone and secondary xylem. We used a sliding microtome (Shenkung Dapples, Mikrot L-serie 1117, Switzerland). The sections were double stained with 0.1% safranin O and fast-green solutions especially good for meristematic tissues (Table 1). The following measurements were performed on each individual according to the methodology of (Delvaux, Sinsin, Van Damme, & Beeckman, 2013). In the wood, the density of the vessels in a surface of 1 mm² was calculated, the diameter of the vessels (on a slide, at least 25 vessels were measured), the area occupied by the vessels, fibers, parenchyma and rays were determined. The number of cells constituting the cambial zone was also determined. These measurements were done in the wood before and after debarking, close to the cambial zone.

Statistical Analysis
All of these semi-automatic measurements were made using the SpectrumSee-F digital image analysis software. Statistical analyzes were carried out using the SPSS software and specifically ANOVA, and the averages were separated by the DUNCAN test.

Anatomical Characteristics of Garcinia lucida Wood
The characterization of the wood of a species is the set of processes consisting in determining the anatomical characteristics of the cells that make up the wood.
Growth rings: 1: They are clearly visible and distinct Vessels: 5: In the vessels, the pores were disseminated or diffuse; Vessels partly solitary, partly in radial multiples of 2-4, or small clusters, absence of features 9, 10, and 11; 13: the type of perforation was simple; 41: the tangential diameter of the vessels was average and varied from 54 to 56 μm ( Figure 1a

Recovery of Wound after Debarking
The recovery rate gradually increased in both species, so they are among the species that heal their wound after debarking (table 2, figure 3).

Characterization of the Cambial Zone
The cambial zone of these two species was composed of a number of variable cells; and these variations were seasonal. For Garcinia lucida, we found that the number of cells constituting the cambial zone in the dry season varies from 06 to 07 cells, whereas in the rainy season it increases and varies from 09 to 10 cells. In the case of Scorodophloeus zenkeri, the same observation was made: in the dry season, the cambial zone was composed of 10 to 11 cells, while in the rainy season it consists of 10 to 12 cells. In G. lucida, although the number of vessels per mm² in the wood before debarking (33 ± 7.98 / mm² ) and regenerated wood (31± 13.83 / mm² ) was different, this density does not differ significantly. With regard to the diameter of the vessels, a significant difference was observed between the initial wood and the regenerated wood. This diameter was higher in the initial wood (56.17± 5.33 μm) compared to the regenerated wood (44.98± 3.52 μm). The areas occupied by parenchyma and fibers were significantly higher in the regenerated wood compared to the initial wood (p <0.05). However, the surfaces occupied by the vessels and the rays remain statistically comparable (p> 0.05) before and after peeling ok bark (table 3).

Modification in Newly Established Wood
In addition to these modifications, there were also other changes in the anatomy of this wood. In cross-section, some tissues were still diffuse in the wood; the diameter of the vessels was small and it was not easy to differentiate or clearly distinguish the different tissues of the wood seen in cross section like parenchyma and wood rays (figure 1d). The density of the vessels in the regenerated wood was changed from the original wood. Moreover, the diameter of the vessels, the surface occupied by the parenchyma and the rays has also undergone modifications. For S. zenkeri, the density of the vessels in the regenerated wood is greater than that of the wood before debarking. It increased from 29 vessels per mm² to 32 vessels per mm² in regenerated wood, but there was no significant difference (table 4). The diameter of the vessels was 63.43 μm for the wood before debarking and 44.39 μm for the regenerated wood; these values are comparable (p> 0.05). The area occupied by the rays is comparable in the wood before and after debarking (p> 0.05). As for parenchyma and fibers, the area they occupy in regenerated wood was significantly higher than that of the wood before bark peeling (p <0.05).
There were also other changes in the anatomy of this wood. In regenerated wood, some tissues were not yet well differentiated, including parenchyma, rays and fibers. It was also observed that the diameters of vessels were very small compared to vessels before debarking (figure 2d).

Discussion
The tree has defence mechanisms based on its ability to compartmentalize wounds. But the effectiveness of these mechanisms is limited in the case of excessive injuries or too severe pruning. Among the practices that result in tree weakening, cuts on large sections and debarking can be cited as an example. The closure of the wound begins with the formation of callus tissue consisting of soft tissue masses of the parenchyma (Fahn, 1985), that gradually invade the wound surface (Bai, Chaney, & Qi, 2005). When the wounds are too numerous and too large, the wood becomes vulnerable and is inevitably infected and colonized by the lignivorous fungi which slowly decompose it. The response of the tree to debarking is first of all specific to each species (Delvaux, Sinsin, Darchambeau, & Van Damme, 2009). After the tree bark was removed from the field, no deaths from the 11 sampled trees were recorded. This could be explained by the debarking on a rectangular area of 10cm wide by 30cm long. This is not a 100% debarking which in fact consists in removing the bark over the entire circumference of the tree, because when debarking is done at 100% it prevents the growth of the tree, because the wound is considerable and damages the tissues of the nutrient pipes of the tree, namely the phloem for nutrients and the xylem for water (Delvaux et al., 2009).
The results obtained in the characterization of the wood were similar to those of the INSIDE WOOD database concerning those species. Changes were observed in the cambial zone of these two species and this according to the seasons. The cambial zone is the part of the tree that is found between the primary xylem and the primary phloem. In the cambial area of the wood, there is an area of cells with little differentiation to active divisions. This generating zone called vascular cambium produces cells that differentiate into secondary xylem and secondary phloem or liber (Dié et al., 2012). In the cambial zone of G. lucida and S. zenkeri, during the rainy season there was an increase in the cells constituting this zone. This could be explained by the favorable climate conditions in this season and the abundance of rainfall. Some studies of the causes of anatomical changes in water supply show that even a short drought during the period of timber formation would result in a corresponding reduction in auxin content, which is a phytohormone inducing cell division (Polge & Keller, 1986). In fact, parenchyma is the most abundant tissue in the regenerated wood; they are made up of globular shaped cells and participate in the nutritive functions of the tree (Evert, 2006).
For the two species studied, regenerated wood vessels were smaller. As regards the density of the vessels, in G. lucida, it is slightly higher in the initial wood than in the regenerated wood. In the wood of S. zenkeri, it is found that the density of the vessels in the regenerated wood is slightly greater than that of the wood before debarking. Similarly, the diameter of the vessels and the area occupied by the vessels in the regenerated wood are smaller than those found in the original wood. Unlike G. lucida, the results obtained in S. zenkeri wood are also similar to those of Delvaux, Sinsin, Van Damme, & Beeckman (2010) who found that during early wound healing, all the species they studied produced vessels with smaller area than in unaffected wood and this significantly decreased the specific conductive area in eight of the investigated species. Small vessels contribute to a safer conducting system and act as adaptive mechanisms to protect the tree (Aloni & Zimmermann, 1983). One of the consequences of the removal of the bark is the obstruction of the conductive vessels which cause the accumulation of auxin. This phenomenon results to increase of differentiation rate of the vessels in the regenerated wood which would lead to more numerous and narrower vessels in this wood (Mwange, Hou, & Cui, 2003;Evert, 2006). In addition, Mwange et al. (2003) also stated that the first steps bark recovery, namely division and differentiation of non-mature cells and the formation of cambium, are dependent on auxin.

Conclusion
In conclusion, Garcinia lucida and Scorodophloeus zenkeri changes in space and time in the anatomy of their wood after wounding; structural changes in the newly established wood are observed, in particular the tissues of regenerated wood such as rays and vessels which are not yet well differentiated as in the wood before wounding; G. lucida regenerates its bark more rapidly than S. zenkeri. In G. lucida, the number of vessels per mm² in the wood before debarking (33 ± 7.98 / mm² ) and regenerated wood (31± 13.83 / mm² ) is different. With regard to the diameter of the vessels, a significant difference is observed between the wood before (56.17± 5.33 μm) and after debarking (44.98± 3.52 μm). The areas occupied by parenchyma and fibers are significantly higher in the regenerated wood compared to the initial wood (p <0.05). For S. zenkeri, the density of the vessels has increased from 29 vessels per mm² to 32 vessels per mm² in regenerated wood, the diameter of the vessels is 63.43 μm for the wood before debarking and 44.39 μm for the regenerated wood. As concerning S. zenkeri, the area occupy by the parenchyma has increased after wounding. For these two species, smaller vessels obtained in regenerated wood contribute to a safer water-conducting system and are an adaptive mechanism. So, vessel diameter is an important parameter for assessing the ascension of water and minerals from roots to leaves and the adaptation of plants to their environment. We can conclude with Delvaux et al. (2010) that vessels appeared to be very good anatomical indicators of the tree's reactions to debarking.